Sage Journals: Discover world-class research

Abstract

Objectives

The aim of this study was to investigate the biological variation of total thyroxine (T4), free T4 (fT4) and thyroid-stimulating hormone (TSH) in 11 clinically healthy cats aged between 3 and 15 years old, in Sydney, Australia.

Methods

Blood was collected weekly for up to 6 weeks and serum T4, fT4 and TSH concentrations were analysed using canine-specific reagents. Restricted maximum likelihood was used to estimate within-subject, between-subject and analytical variance components, which were recorded in terms of the related coefficients of variation. The index of individuality and reference change values were then calculated for each analyte.

Results

T4 and TSH had intermediate individuality, indicating both subject-based and population-based reference intervals (RIs) could be used, with the knowledge that population-based RIs are suboptimally sensitive. fT4 had high individuality, indicating subject-based RIs are more appropriate than population-based RIs.

Conclusions and relevance

This study has demonstrated that subject-based RIs could be more sensitive than population-based RIs for the diagnosis of thyroid dysfunction in cats.

Keywords

Thyroid hyperthyroidism biological variation reference interval

Introduction

Feline hyperthyroidism is an increasingly common disease in older cats,^1,2 and is most commonly diagnosed by measuring serum total thyroxine (T4) and free T4 (fT4) concentrations, in conjunction with a suite of clinical signs. However, 10% of cats with clinically overt hyperthyroidism and 40% of cats with subclinical hyperthyroidism have a serum T4 value within the current population-based reference interval (RI).³ The majority of cats that become hyperthyroid have a 1–3 year subclinical phase characterised by low serum thyroid-stimulating hormone (TSH) and normal serum T4 concentrations.⁴ Therefore, a more sensitive measure of thyroid function would facilitate earlier diagnosis and management of disease, thereby reducing the clinical impact of hyperthyroidism and improving patient outcomes.

Traditionally, population-based RIs have been used in veterinary medicine, formulated by measuring analytes in a population of healthy individuals followed by statistical development of RIs.^5,6 However, for some analytes, where variation between individuals in a population exceeds variation within an individual over time, clinically significant changes in analyte levels may be masked when population-based RIs are used.^5,7–9 Alternatively, subject-based RIs monitor an individual’s change in analyte value over time, and can be more sensitive when there is a wide variation in normal values between individuals. This has been demonstrated to be the case in cats for alkaline phosphatase, alanine aminotransferase, creatinine, cholesterol, total protein,^10–12 albumin, N-terminal pro-brain natriuretic peptide,¹³ insulin-like growth factor,¹⁴ T4,^12,15 and fT4 and TSH.¹⁵

In humans, the ratio of within- to between-individual variation is low for TSH, T4 and triiodothyronine (T3), indicating that population-based RIs are relatively insensitive to aberrations from normality in the individual.^16–19 In addition, low sensitivity of population-based RIs for T4 and T3 contributes to under-diagnosis of thyroid disease.¹⁷ Inter-individual variation of TSH in dogs is significantly greater than intra-individual variation,²⁰ and inter-individual variation in T4 in cats is significantly greater than intra-individual variation.¹² This suggests a similar lack of sensitivity of population-based RIs for the diagnosis of feline thyroid disease.

Biological variation refers to physiological variation in analyte concentration, and consists of cyclical, physiological and inherent variation. Cyclical variation can be seasonal, monthly or circadian, and physiological variation may be associated with age, sex, breed and pregnancy/reproductive status, or affected by geographical location.^9,21 To assess the utility of population- and subject-based RIs for a particular analyte, inherent biological variation of an analyte (CV_I) and variation between individuals (CV_G) must be determined, and from these values, an index of individuality (II) is determined.^7,9,21,22 A reference change value (RCV) can then be calculated, providing a reference for the degree of change in a given analyte over time, which can be applied in a clinical setting.^7,9,21 This study aimed to further investigate the use of subject-based RIs in the diagnosis of feline hyperthyroidism, by determining the CV_I, CV_G, II and RCV for T4, fT4 and TSH.

Materials and methods

Animals

Eleven clinically healthy client-owned cats were included in the study, subject to the following inclusion criteria: >2 years of age, no history of major illness, thyroid disease, or current or recent medication other than prophylactic treatments, and not gravid. Initial health status was based on physical examination, owner interview, and routine biochemistry and haematological analysis. Physical examination included cardiac and pulmonary auscultation, femoral pulse rate, assessment of mucous membrane colour and capillary refill time, thyroid palpation, abdominal palpation, determination of body condition score (BCS) and assessment of demeanour. Ongoing health status during the 6-week study period was based on owner interview, cat demeanour and basic physical examination.

Twelve cats were originally included in the study; one female cat was excluded after repeatedly returning elevated serum T4 and low serum TSH concentrations, consistent with hyperthyroidism. One cat was withdrawn after three blood collections; however, data from these collections were included. Of the 11 cats included, there were seven domestic shorthairs, two domestic longhairs, one Devon Rex and one Persian crossbreed. The cats were between 3 and 15 years of age (mean 8.3 years, median 8.0 years) with six males and five females. BCSs ranged between 2.5 and 4 (mean 2.75, median 2.5) on a scale of 1–5, where 1 is emaciated and 5 is obese.²³ The study had ethical approval from the University of Sydney Animal Ethics Committee (2015/847).

Sample collection and handling

Samples were collected between May and November 2017. Blood was collected from each cat once a week for 6 weeks (n = 4), 5 weeks (n = 6) or 3 weeks (n = 1). Samples were collected within 24 h of a 1-week interval, at approximately the same time of day. Up to 3 ml of venous blood was collected from the jugular or cephalic vein using a 21–23 G needle (BD PrecisionGlide) with a 3 or 5 ml syringe attached on first sampling; up to 1 ml of blood was collected for subsequent sampling events. Blood was collected into 1.3 ml plain serum microtubes (Sarstedt) and centrifuged at 1790 g for 5 mins within 1–6 h of collection. Serum was pipetted into 1.2 ml cryovials (Biosigma) and frozen immediately at –80°C and stored for up to 8 months. While the fasting of cats was requested prior to sample collection, both fasted and non-fasted sampling occurred. Multiple samples from one cat were grossly lipaemic, and serum from these samples was centrifuged for an additional 5 mins at 1790 g and accumulated lipid removed prior to frozen storage.

Laboratory methods

T4 and fT4 were run on an automated immunoassay system (Immulite 1000; Siemens Healthcare), using a solid-phase, competitive chemiluminescent enzyme immunoassay (CEIA); (Immulite 1000 Canine Total T4 [PILKCT-13, 2015-09-03] and Veterinary Free T4 [PILVF4-2, 2017-03-06]; Siemens Healthcare). TSH was measured on the same system using a solid-phase, two-site chemiluminescent immunometric assay (Immulite 1000 Canine TSH [PILKKT-12, 2017-03-06]; Siemens Healthcare). Each reagent was calibrated with assay-specific adjusters before the samples were analysed. Manufacturer controls for T4, fT4 and TSH (Immulite 1000 Canine controls; K9CON) were analysed prior to analysis of the study samples, as well as once during sample analysis, with all controls within acceptable limits.

All samples were analysed on the same day to reduce intra- and inter-run variation. Samples were thawed at room temperature and each sample was then analysed in duplicate, by a single operator using a single lot of T4, fT4 and TSH reagents.

Statistical analysis

All statistical analyses were performed in R 3.5.1 (R Core Team 2018).²⁴ Values below the detection limit were entered as half the detection limit to facilitate statistical analysis.

Estimation of variance

Restricted maximum likelihood was used to estimate within-subject, between-subject and analytical variance components, which were recorded in terms of the related coefficients of variation (CV_I, CV_G and CV_A, respectively). For each of the analytes (T4, fT4, TSH), patient ID was specified as a random variable with sample week nested within patient ID. Model residuals were checked for normality using histograms and quantile-quantile plots.

Identification of outliers

Identification of outlying variances was performed by visual inspection of the sample distribution and by Tukey’s criterion, as outlined by Horn et al.²⁵ Outliers for each analyte were assessed across the entire sample, as well as within each subject individually. Where one of two duplicates for a given patient/week was identified as an outlier, both duplicates were removed on the assumption that the differences between the duplicates was due to an unacceptable level of analytical variation or analytical error. Where both duplicates for a given patient/week were identified as outliers, the data were retained on the assumption that the variation represented biological rather than analytical variation. This resulted in two sample duplicates for analytes T4 from one cat being identified as subject-based outliers, and subsequent removal from the statistical analysis. In addition, one cat repeatedly returned serum T4 concentrations >80 nmol/l, with all TSH values below the detection limit. This cat was excluded from the study, meeting the exclusion criteria of evidence of thyroid disease.^4,26

Estimation and interpretation of index of individuality (II)

II was estimated based on the following definition:

I I = \sqrt{\frac{C V_{I}^{2} + C V_{A}^{2}}{C V_{G}}} .

Interpretation of II was based on the following.

II ⩽0.6 (high II)

Subject-based RIs are recommended, while population-based RIs are of limited use.

II = 0.6–1.4 (intermediate II)

Subject-based reference intervals can be used as for high II. Population-based reference values can be used; changes may not be detected as well as for low II.

II ⩾1.4 (low II)

Population-based RIs are appropriate. No additional information is likely to be gained from subject-based RIs.^7,9,27

Analytical quality specifications were drawn from Peterson et al,²⁸ with analytical imprecision categorised as minimum, desirable and optimal, calculated as CV_MIN <0.75 CV_I, CV_DES <0.50 CV_I and CV_OPT <0.25 CV_I, respectively.²⁸

Calculation of RCVs

RCVs for 95% confidence intervals were calculated using the following:

R C V = Z \sqrt{2} \sqrt{C V_{I}^{2} + C V_{A}^{2}} .

Uni- and bi-directional changes were calculated using (Z = 1.65) and (Z = 1.96), respectively.

Results

For TSH, 8/56 (14.3%) results were below the detection limit, so a value half the value of the detection limit (0.15 µg/l) was substituted for statistical analysis. Analytical precision of all analytes was adequate, with CV_A ⩽0.75 CV_I in all cases.

TSH and T4 CV_A values were within the desirable range (CV_A ⩽0.50 CV_I), and TSH had a CV_A value within the optimum range (CV_A ⩽0.25 CV_I). The calculated II for T4 and TSH was intermediate (between 0.6 and 1.4), while the calculated II for fT4 was high (0.58; Table 1).

Table 1

Mean concentration and range of observed values, CV_I, CV_G, CV_A, index of individuality (II), and one- and two-sided reference change values (RCVs) for thyroxine (T4), free T4 (fT4) and thyroid-stimulating hormone (TSH) in clinically healthy cats (n = 11) sampled weekly over a 6-week period

Analyte	Mean (range)	CV_I %	CV_G %	CV_A %	II	RCV %(one sided)	RCV %(two sided)
T4 (nmol/l)	23.0 (13.8–32.2)	11.6	12.7	4.2	0.98	29	34
fT4 (pmol/l)	11.7 (1.93–20.3)	14.9	31.8	10.8	0.58	43	51
TSH (µg/l)	0.051 (0.015–0.124)	32.3	28.7	3.7	1.13	76	90

CV_I = inherent biological variation of an analyte; CV_G = variation between individuals; CV_A = analytical variation inherent to a test

Discussion

There is increasing interest in the use of subject-based RIs as an alternative to population-based RIs in veterinary medicine. This study provides data regarding biological variation of the analytes most commonly used for diagnosing feline thyroid disease.

CV_A represents the analytical variation, otherwise known as the random error or imprecision, inherent to a particular test.⁹ When assessing biological variation, it is essential to evaluate CV_A, as it represents an increase in CV_I that is not attributable to true intra-individual variation. CV_A ⩽0.50 CV_I is widely accepted as the desirable maximum analytical variation,^9,21,29 although Strage et al¹⁴ suggest CV_MIN <0.75 CV_I, CV_DES <0.50 CV_I and CV_OPT <0.25 CV_I (minimum, desirable and optimum performance, respectively), acknowledging that some assays currently in use lack the desired analytical precision.

Applying these standards to the current data set, the analytical precision of all analytes was adequate, with CV_A ⩽0.75 CV_I in all cases. With analytical variation of T4 within the desirable range, variation in results can be attributed to sample variation with reasonable confidence. While fT4 demonstrated acceptable analytical variation, it was above the desirable range, meaning variation is less confidently attributed to inherent variation of the sample. While the CV_A for TSH satisfied the optimal criteria, the use of a value of half the minimum value most likely artefactually decreased CV_A, making the analytical goals easier to achieve.

Calculated II for T4 and TSH was intermediate, indicating that both subject-based and population-based RIs can be used, although population-based RIs should be used with caution. This means that comparing a single result with a population-based RI will result in the detection of some values as lying outside the RI; however, clinically significant changes could be masked in animals that have a lower homeostatic set point. This is reflected in the observation that 10% of clinically hyperthyroid cats have a T4 value within the normal RI.³ For these animals, applying an RCV to two samples taken at least 1 week apart could result in an earlier diagnosis of hyperthyroidism than waiting until the T4 value is above the population-based maximum value.

RCVs could also be used to monitor the response to treatment in hyperthyroid cats, by assessing the significance of changes in serum T4 concentration over time. However, as this study only included data from healthy cats, future studies should investigate whether biological variation of these analytes is consistent in hyperthyroid cats.

Biological variation has previously been reported for T4, fT4 and TSH in cats.^12,15 In the current study, and that published by Prieto et al,¹⁵ the II of T4 was considered intermediate, while Falkenö et al¹² reported an II for T4 just within the II range considered ‘high’. The CV_I is similar across all studies (11.6%, 9.0% and 11.4%, respectively), and the greater CV_G (20.0%) reported by Falkenö et al¹² when compared with the current study and Prieto et al (both 12.7%)¹⁵ accounts for this difference.

There are several possible reasons for this difference in CV_G between the studies. Undiagnosed disease (eg, subclinical thyroid disease or non-thyroidal illness) could result in increased or decreased T4 levels within an individual. While there is little information regarding breed differences in thyroid hormone levels in cats, significant differences in normal thyroid hormone concentration among canine breeds have been well documented.^30–33 A variety of feline breeds were included in all three studies, in addition to domestic short- and longhairs. Falkenö et al¹² included a Balinese, a Bengal and a mixed-breed Norwegian Forest Cat; in the current study, a Devon Rex and a Persian crossbreed were included.

Photoperiod and temperature, due to sampling at different geographical locations and time of year, could also have an effect on thyroid hormone levels,^34–37 although these effects are not well characterised in domestic cats. In addition, serum was used in the current study, while Falkenö et al¹² used heparinised plasma, although – as the manufacturer guidelines state that in human samples – the reported mean is 3.68% higher when plasma is used, compared with serum for this assay, which may be more likely to result in a systematic difference.

While there is a slight difference in interpretation of II values derived in the current study and those derived by Falkenö et al,¹² there is agreement in the conclusion that population-based RIs alone are not always sensitive enough when interpreting T4 values in feline patients.

Calculated II for fT4 in the current study was high, suggesting subject-based RIs are recommended, and population-based RIs are of limited use. Prieto et al¹⁵ reported an II within the intermediate range for fT4. In the current study, CEIA was used for measurement of fT4, while Prieto et al¹⁵ used modified equilibrium dialysis (MED). The latter method is considered the gold standard for determination of fT4 in cats, and while fT4 CEIA has been validated for use in healthy cats,^38,39 this difference in method could account for the differing results. fT4 is the unbound, biologically active fraction in serum,^2,38 with <1% of T4 circulating as fT4.² While MED physically separates protein-bound and unbound hormone,⁴⁰ fT4 measurement by CEIA could be more affected by aberrations in binding kinetics of T4 and proteins such as thyroxine-binding pre-albumin, and – to a lesser extent – albumin.^33,41 Factors known to affect binding kinetics, such as non-thyroidal disease and use of certain drugs,⁴¹ were minimised in the current study, although it is possible that undiagnosed disease or analytical factors may have contributed to differences between fT4 measurements and derived values in the two studies. However, the findings of the current study support the use of RCVs, at least in conjunction with population-based RIs, to improve the detection of significant changes in fT4.

TSH is a very useful analyte used in the early detection of thyroid disease in humans and other species, as TSH can be suppressed before elevations in T4 are present.⁴ However, there is no feline-specific TSH reagent currently available, and the insensitive detection limit of canine TSH is a major limitation to its use in feline patients, as it is unable to distinguish between low normal values and values that are suppressed due to early hyperthyroidism.⁴ In this study, the II of TSH is intermediate, compared with an II in the high range reported by Prieto et al.¹⁵ This is largely due to a higher CV_G in that study, and reasons for an increased CV_G, such as undiagnosed disease, breed differences, and effects of photoperiod and temperature, as discussed above, could apply.

It is important to note the necessary use of a single value for each value below the minimum limit in both studies, potentially masking individual variation (CV_I), as well as limiting group variation (CV_G); differences between the number of substituted values could have contributed to the difference in derived values. Owing to this statistical necessity, values derived in this study and those published by Prieto et al¹⁵ are unlikely to be a true reflection of biological variation of TSH in cats.

The development of a feline-specific TSH assay with a suitable detection limit would allow greater precision in the calculation of these values, providing a more accurate understanding of feline TSH variation and the suitability of current RIs.

Conclusions

This study demonstrates the inherent biological variation of serum concentrations of T4, fT4 and TSH in cats. The calculated II for fT4 was high, indicating subject-based RIs are recommended over currently used population-based RIs.

An intermediate II was obtained for T4 and TSH, indicating that both subject-based and population-based RIs can be used to detect deviation from normal values, the latter with the knowledge that there is suboptimal sensitivity. As these results suggest population-based RIs alone are not sufficiently sensitive to interpret T4 values in cats, the use of subject-based RIs and RCVs in conjunction with population-based RIs should be considered when diagnosing and monitoring thyroid disease in cats.

Footnotes

Acknowledgements

The authors would like to thank all the feline participants and their owners, Sydney School of Veterinary Science and Veterinary Pathology Diagnostic Service for laboratory access and Christine Black for her technical support.

Accepted: 5 October 2020

Conflict of interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

Funding for this project was provided by Sydney School of Veterinary Science and Veterinary Pathology Diagnostic Service, University of Sydney.

Ethical approval

This work involved the use of non-experimental animals (owned or unowned) and procedures that differed from established internationally recognised high standards (‘best practice’) of veterinary clinical care for the individual patient. The study therefore had ethical approval from an established committee as stated in the manuscript.

Informed consent

Informed consent (either verbal or written) was obtained from the owner or legal custodian of all animal(s) described in this work (either experimental or non-experimental animals) for the procedure(s) undertaken (either prospective or retrospective studies). No animals or humans are identifiable within this publication, and therefore additional informed consent for publication was not required.

ORCID iD

Anne Jordan

References

Edinboro

Scott-Moncrieff

Janovitz

, et al. Epidemiologic study of relationships between consumption of commercial canned food and risk of hyperthyroidism in cats. J Am Vet Med Assoc 2004; 224: 879–886.

Scott-Moncrieff

JC.

Feline hyperthyroidism. In: Feldman

Nelson

Reusch

, et al (eds). Canine and feline endocrinology. 4th ed. St Louis, MO: Saunders, 2014, pp 136–195.

Peterson

ME.

More than just T4: diagnostic testing for hyperthyroidism in cats.

J Feline Med Surg 2013; 15: 765–777.

Wakeling

Practice tips: use of thyroid stimulating hormone (TSH) in cats.

Can Vet J 2010; 51: 33–34.

Friedrichs

Harr

Freeman

, et al. ASVCP reference interval guidelines: determination of de novo reference intervals in veterinary species and other related topics. Vet Clin Pathol 2012; 41: 441–453.

Geffré

Friedrichs

Harr

, et al. Reference values: a review. Vet Clin Pathol 2009; 38: 288–298.

Fraser

CG.

Inherent biological variation and reference values.

Clin Chem 2004; 42: 758–764.

Iglesias

Cava

Simo

, et al. Within-subject biological variation in disease: collated data and clinical consequences. Ann Clin Biochem 2007; 44: 343–352.

Walton

RM.

Subject-based reference values: biological variation, individuality, and reference change values.

Vet Clin Pathol 2012; 41: 175–181.

10.

Baral

Dhand

Freeman

, et al. Biological variation and reference change values of feline plasma biochemistry analytes. J Feline Med Surg 2013; 16: 317–325.

11.

Trumel

Monzali

Geffré

, et al. Hematologic and biochemical biologic variation in laboratory cats. J Am Assoc Lab Anim Sci 2016; 55: 503–509.

12.

Falkenö

Hillström

von Brömssen

, et al. Biological variation of 20 analytes measured in serum from clinically healthy domestic cats. J Vet Diagn Invest 2016; 28: 699–704.

13.

Harris

Estrada

Gallagher

, et al. Biologic variability of N-terminal pro-brain natriuretic peptide in adult healthy cats. J Feline Med Surg 2017; 9: 216–223.

14.

Strage

Theodorsson

Ström Holst

, et al. Insulin-like growth factor I in cats: validation of an enzyme-linked immunosorbent assay and determination of biologic variation. Vet Clin Pathol 2015; 44: 542–551.

15.

Prieto

Carneya

Millera

, et al. Short-term biological variation of serum thyroid hormones concentrations in clinically healthy cats. Domest Anim Endocrinol 2020; 71: 106389. DOI: 10.1016/j.domaniend.2019.106389.

16.

Andersen

Bruun

Pedersen

, et al. Biologic variation is important for interpretation of thyroid function tests. Thyroid 2003; 13: 1069–1078.

17.

Andersen

Pedersen

Bruun

, et al. Narrow individual variations in serum T(4) and T(3) in normal subjects: a clue to the understanding of subclinical thyroid disease. J Clin Endocrinol Metab 2002; 87: 1068–1072.

18.

Browning

Ford

Callaghan

SJ.

Intra-and inter-individual biological variation of five analytes used in assessing thyroid function: implications for necessary standards of performance and the interpretation of results.

Clin Chem 1986; 32: 962–966.

19.

Maes

Mommen

Hendrickx

, et al. Components of biological variation, including seasonality, in blood concentrations of TSH, TT3, FT4, PRL, cortisol and testosterone in healthy volunteers. Clin Endocrinol 1997; 46: 587–598.

20.

Iversen

Jensen

Høier

, et al. Biological variation of canine serum thyrotropin (TSH) concentration. Vet Clin Pathol 1999; 28: 16–19.

21.

Fraser

Harris

EK.

Generation and application of data on biological variation in clinical chemistry.

Crit Rev Clin Lab Sci 1989; 27: 409–437.

22.

Harris

Kanofsky

Shakarji

, et al. Biological and analytical components of variation in long-term studies of serum constituents in normal subjects; II estimating biological components of variation. Clin Chem 1970; 16: 1022–1027.

23.

Witzel

Kirk

Henry

, et al. Use of a morphometric method and body fat index system for estimation of body composition in overweight and obese cats. J Am Vet Med Assoc 2014; 244: 1285–1290.

24.

R Core Team.

R: a language and environment for statistical computing.

Vienna: R Foundation for Statistical Computing, 2018.

25.

Horn

Feng

, et al. Effect of outliers and nonhealthy individuals on reference interval estimation. Clin Chem 2001; 47: 2137–2145.

26.

Peterson

Melian

Nichols

Measurement of serum concentrations of free thyroxine, total thyroxine, and total triiodothyronine in cats with hyperthyroidism and cats with nonthyroidal disease.

J Am Vet Med Assoc 2001; 218: 529–536.

27.

Harris

EK.

Effects of Intra- and inter-individual variation on the appropriate use of normal ranges.

Clin Chem 1974; 20: 1535–1542.

28.

Petersen

Fraser

Jerqensen

, et al. Combination of analytical quality specifications based on biological within- and between-subject variation. Ann Clin Biochem 2002; 39: 543–550.

29.

Cotlove

Harris

Williams

GZ.

Biological and analytic components of variation in long-term studies of serum constituents in normal subjects. 3. Physiological and medical implications.

Clin Chem 1970; 16: 1028–1032.

30.

Van Geffen

Bavegems

Duchateau

, et al. Serum thyroid hormone concentrations and thyroglobulin autoantibodies in trained and non-trained healthy whippets. Vet J 2006; 172: 135–140.

31.

Panacova

Kocj

Kolb

, et al. Thyroid testing in Sloughis. J Vet Intern Med 2008; 22: 1144–1148.

32.

Seavers

Snow

Mason

, et al. Evaluation of the thyroid status of Basenji dogs in Australia. Aust Vet J 2008; 86: 429–434.

33.

Scott-Moncrieff

JC.

Hypothyroidism. In: Feldman

Nelson

Reusch

, et al (eds). Canine and feline endocrinology. 4th ed. St Louis, MO: Saunders, 2014, pp 77–135.

34.

Oohashi

Yagi

Uzuka

, et al. Seasonal changes in serum total thyroxine, free thyroxine, and canine thyroid-stimulating hormone in clinically healthy beagles in Hokkaido. J Vet Med Sci 2001; 63: 1241–1243.

35.

Tomasi

TE.

Utilization rates of thyroid hormones in mammals.

Comp Biochem Physiol Part A Physiol 1991; 100: 503–516.

36.

Tuckova

Fialkovicova

Baranova

, et al. Seasonal effect on serum levels of thyroid hormones in German Shepherds and Dachshunds and their health status. Vet Med (Praha) 1995; 40: 249–252.

37.

Byers

Hunter

Seal

, et al. Effect of season on seminal traits and serum hormone concentrations in captive male Siberian tigers (Panthera tigris). J Reprod Fert 1990; 90: 119–125.

38.

Peterson

Broome

Robertson

Accuracy of serum free thyroxine concentrations determined by a new veterinary chemiluminescent immunoassay in euthyroid and hyperthyroid cats. Proceedings of the European College of Veterinary Internal Medicine (ECVIM); 2011 Jan 8–10; Seville, Spain.

39.

Stammeleer

Buresova

Stock

, et al. Comparison of free thyroxine measurement by chemiluminescence and equilibrium dialysis following 131I therapy in hyperthyroid cats. J Feline Med Surg 2020; 22: 1114–1120.

40.

Thienpont

Van Uytfanghe

Velkeniers

Determination of free thyroid hormones.

Best Pract Res Clin Endocrinol 2013; 27: 689–700.

41.

Schachter

Nelson

Scott-Moncrieff

, et al. Comparison of serum-free thyroxine concentrations determined by standard equilibrium dialysis, modified equilibrium dialysis, and 5 radio-immunoassays in dogs. J Vet Intern Med 2004; 18: 259–264.

Biological variation of total thyroxine (T4),free T4 and thyroid-stimulating hormone in 11 clinically healthy cats

Abstract

Objectives

Methods

Results

Conclusions and relevance

Keywords

Introduction

Materials and methods

Animals

Sample collection and handling

Laboratory methods

Statistical analysis

Estimation of variance

Identification of outliers

Estimation and interpretation of index of individuality (II)

II ⩽0.6 (high II)

II = 0.6–1.4 (intermediate II)

II ⩾1.4 (low II)

Calculation of RCVs

Results

Discussion

Conclusions

Footnotes

Acknowledgements

Conflict of interest

Funding

Ethical approval

Informed consent

ORCID iD

References