Effect of EDTA on measurement of cortisol and thyroxine by chemiluminescent enzyme immunoassay in dogs

Validation of a fully automated chemiluminescent immunoassay for cattle serum and plasma progesterone measurement

Introduction

Monitoring circulating progesterone concentration ([P4]) is an important component of basic and applied reproduction research and clinical settings. IMMULITE^® 2000 XPi (Siemens Healthineers, Cary, NC) is a newly upgraded fully automated immunoassay system marketed for human use to measure concentrations of different measurands including P4.

Objectives

Our objective was therefore to characterize the analytical performance of the IMMULITE^® 2000 XPi P4 immunoassay (IPI) across the reportable range in serum and plasma of cattle.

Methods

The IPI validation protocols included characterization of the method linearity, within-run, and between-run precision through calculation of the coefficient of variation (CV). The method accuracy was assessed through the calculation of the spiking-recovery (SR) bias across the reportable range (0.2-40.0 ng/mL). Passing-Bablok regression and Bland-Altman plots were used to determine the interlaboratory bias for two laboratories. Three types of observed total error (TEo) were calculated based on the considered type of bias, TEo_SR (spiking-recovery), TEo_RB (range-based bias), and TEo_AB (average-based bias).

Results

The IPI was linearly related to the true value (R ² = 0.997) across the reportable range. The within-run and between-run precision (CV%) of the IPI for both serum and plasma [P4] of clinical relevance (1, 2, 5, and 10 ng/mL) were <5 and <10%, respectively. The TEo reported here for serum and plasma at [P4] of 1 and 5 ng/mL was ~20 and 25%, respectively. Of interest, the three types of TEo were relatively similar regardless of the considered bias.

Conclusions

We concluded that the automated IPI provides a precise, accurate, reliable, and safe method for measuring [P4] in both serum and plasma of cattle. Consistent with the manufacturer's recommendations, the serum matrix is more accurate than plasma.

Effects of EDTA on chemiluminescent immunoassay measurement of ACTH, cortisol, and thyroid hormones in dogs

When measuring blood hormones, pre-analytical sample handling can impact the quality of the results. Previous studies have shown improved stability of canine cortisol in ethylenediaminetetraacetic acid (EDTA) plasma compared to serum and interchangeability of serum and plasma when cortisol is measured by radioimmunoassay. Additionally, cortisol samples were also interchangeable when measured by chemiluminescent immunoassay if the EDTA concentration was consistent with that of optimally filled tubes, whereas excess EDTA interfered with the measurement of cortisol and serum and EDTA plasma were not interchangeable when measuring total thyroxine (TT4). The main limitation of these studies was that they were performed by spiking pooled serum samples with EDTA or in previously collected blood samples submitted to a clinical pathology laboratory. The purpose of the present study was to evaluate the effect of EDTA on the measurement of adrenocorticotropic hormone (ACTH), cortisol, TT4, free thyroxine (FT4), and thyroid stimulating hormone (TSH) in healthy dogs using the Siemens IMMULITE 1000. Whole blood from forty dogs was aliquoted into three Monoject sample tubes: no additive, completely filled EDTA tube, and 50% filled EDTA tube. Handling and storage conditions were identical, and all samples were analyzed on the same day. Bland-Altman plots and Passing-Bablok regression were used to assess agreement and risks for error, respectively. Proportional errors were found between serum and plasma samples for ACTH, cortisol, TT4, FT4, and TSH; systematic errors were also found for FT4. There was poor agreement and clinically significant differences between the measured concentrations of all hormones in serum and plasma, proving that these sample types are not interchangeable. Incompletely filled EDTA tubes were associated with significantly lower ACTH concentrations compared to completely filled EDTA tubes. When measured by chemiluminescent immunoassays that utilize alkaline phosphatase at the reporter enzyme, serum should be used for cortisol, TT4, FT4, and TSH, while plasma from completely filled EDTA tubes should be used for ACTH.

Frequency and risk factors for naturally occurring Cushing's syndrome in dogs attending UK primary-care practices

Objectives

To estimate the frequency and risk factors for Cushing's syndrome in dogs under UK primary veterinary care.

Materials and Methods

Dogs with Cushing's syndrome were identified by searching electronic patient records of primary‐care veterinary practices. Pre‐existing and incident cases of Cushing's syndrome during 2016 were included to estimate the 1‐year period prevalence. Incident cases were used to estimate the annual incidence and to identify demographic risk factors for the diagnosis of Cushing's syndrome in dogs, through multivariable logistic regression.

Results

Analysis included 970 pre‐existing and 557 incident cases of Cushing's syndrome from a population of 905,544 dogs. The estimated 1‐year period prevalence for Cushing's syndrome in dogs under veterinary care was 0.17% (95% confidence interval 0.16 to 0.18) and incidence was 0.06% (95% confidence interval 0.05 to 0.07). In multivariable logistic regression modelling, the Bichon frise (odds ratio=6.17, 95% confidence interval 4.22 to 9.00), Border terrier (5.40, 95% confidence interval 3.66 to 7.97) and Miniature schnauzer (3.05, 95% confidence interval 1.67 to 5.57) had the highest odds of Cushing's syndrome. The Golden retriever (0.24, 95% confidence interval 0.06 to 0.98) and Labrador retriever (0.30, 95% confidence interval 0.17 to 0.54) were the most protected breeds. Increasing age, bodyweight greater than the breed‐sex mean and being insured also showed increased odds of Cushing's syndrome.

Clinical Significance

As Cushing's syndrome is predominately diagnosed and managed in primary‐care practice, this study provides valuable new information of its epidemiology in this setting. Demographics reported are supportive of previous work and additional novel associations identified, such as the Border terrier, could enhance the index of suspicion for veterinarians.