Abstract
We investigated the stability of adrenocorticotropic hormone (ACTH) in plasma after freezing for different lengths of time. The plasma ACTH concentrations of 12 horses were measured on day 0 (baseline) and over time, after stimulation with thyrotropin-releasing hormone. Samples were stored at −80°C for 3, 7, 30, 60, and 90 d, or at −20°C for 3, 7, 30, and 60 d, or between ice packs at −20°C for 3 and 7 d prior to determination of ACTH concentration. ACTH concentrations were compared to baseline (non-frozen day 0 plasma) for each storage method using a mixed model with repeated measures in which each horse served as its own control and day was the repeated effect. Statistical significance was set at p ≤ 0.05, and 0.05 < p < 0.10 was considered a trend. Plasma ACTH frozen at −20°C or at −80°C resulted in degradation of ACTH compared to baseline samples at 60 and 90 d respectively. There was no degradation of ACTH after 7 d when stored between ice packs, or before 30 d at −20°C, or before 60 d at −80°C.
Measurement of adrenocorticotropic hormone (ACTH) has become the most common laboratory method utilized in the diagnosis of pituitary pars intermedia dysfunction (PPID) in horses. 1 ACTH has been reported to be fragile in whole blood samples, and is affected by heat and time prior to centrifugation.6,7 Human ACTH concentrations have been shown to be stable for 4 h at room temperature 2 and significantly decreased by 72 h. 8 Increased time from collection to centrifugation also results in decreased measured human ACTH concentrations. 6 Equine ACTH has been shown to be stable without centrifugation for up to 8 h stored at 21 or 4°C, and was stable in plasma stored at −20 and −80°C for 30 d. 5 Consequently, recommendations have been to refrigerate whole blood samples prior to centrifugation and to separate plasma within 8 h of collection. 4 If plasma samples cannot be shipped the day of collection to the laboratory, plasma should be frozen until shipment. In a clinical setting, samples are unlikely to be held for more than a few days in a standard kitchen refrigerator. For research purposes, samples may be held for months at −80°C. In order to understand whether freezing has any effects on equine ACTH concentration, we studied the stability of equine ACTH in plasma after freezing for different lengths of time at various temperatures.
Our study was approved by the Middle Tennessee State University (MTSU) Animal Care and Use Committee via protocol 17-2013. In May, 12 horses (5 mares, 6 geldings, and 1 stallion), 14–29 y old, from the MTSU herd were screened for ACTH with levels of 3–14 pmol/L (<7 pmol/L = negative, >11 pmol/L = positive for PPID). In September, 1.0 mg of thyrotropin-releasing hormone suspended in 1 mL of saline was administered intravenously to each of the same 12 horses. Ten min later, blood samples were collected in EDTA tubes and refrigerated at 5°C until centrifuged at 1,000 × g for 10 min within 2 h of collection. EDTA is the recommended anticoagulant given the greatly decreased ACTH concentrations obtained when heparin is used as an anticoagulant. 3 Plasma was stored in microcentrifuge tubes and frozen for various lengths of time and conditions. Day 0 samples were frozen at −80°C and shipped that day overnight with cold packs to the Animal Health Diagnostic Center at Cornell University (Ithaca, NY) for ACTH analysis. A set of 12 plasma samples stored at −80°C was shipped on 3, 7, 30, 60, and 90 d post-collection. Another set of 12 samples was stored at −20°C and shipped on 3, 7, 30, and 60 d post-collection. Another set of 12 samples was stored between ice packs to mitigate fluctuation of temperature as a result of opening and closing of the freezer door at −20°C, and was sent overnight on days 3 and 7 post-collection.
Samples were batch analyzed for plasma ACTH by a chemiluminescent immunoassay validated previously for horses. 4 The immunologic specificity was 99.1%, accuracy was 101%, mean intra-assay %CV was 9.3%, inter-assay CVs was 9.1% for low control (mean = 6 pmol/L) and 7.0% for high control (mean = 84 pmol/L)], and sensitivity for this method was 2 pmol/L. Sensitivity and specificity of ACTH levels for detecting PPID for this assay was reportedly 84% (95% confidence interval [CI]: 60, 97) and 78% (95% CI: 63, 88), respectively. 4
Within each storage method, ACTH concentrations were compared to baseline using a mixed model with repeated measures (v.9.2; SAS Institute, Cary, NC) in which each horse served as its own control, with day as the repeated effect. Pearson correlations were used to evaluate the relationship between ACTH concentrations and percent change from baseline (%CFB). Statistical significance was set at p ≤ 0.05, and 0.05 < p < 0.10 was considered a trend. Data were summarized using means, standard errors, ranges, and %CFB.
–80°C freezer: Mean baseline ACTH concentration for plasma stored at −80°C was 86 pmol/L, and declined 6.9% to a low of 80 pmol/L by day 90 (p = 0.05). Through 60 d, the %CFB never varied > 2% and was not different (p > 0.62) from baseline. At 90 d, the %CFB was −6.9% and different from baseline (p = 0.03; Table 1). The ACTH concentrations in plasma stored at −80°C were 10–265 pmol/L on day 0 and 10–239 pmol/L at 90 d. Correlations between ACTH concentration and %CFB were not observed (R = −0.032; p = 0.79), thus degradation was similar across concentrations.
Adrenocorticotropic hormone concentrations (pmol/L)* from 12 equine plasma samples stored under different conditions.
%CFB = percent change from baseline; SE = standard error within horse.
Divide pmol/L by 0.22 to convert to pg/mL.
Significant at p = 0.03.
–20°C freezer: Mean baseline ACTH concentration for plasma stored at −20°C was 86 pmol/L, and declined 5.3% to 82 pmol/L by day 60. The day 30 sample was not different from day 0 (p = 0.16) but it was higher than day 7 and day 60 (p = 0.01). Although the ACTH concentrations at baseline and 60 d were not different (p = 0.18), the %CFB tended to be lower than baseline at 7 d (p = 0.08) and was lower by 60 d (p = 0.04; Table 1). The ACTH concentrations in plasma stored at −20°C were 10–265 pmol/L on day 0 and 9–232 pmol/L by 60 d. Correlations between ACTH concentration and %CFB were not observed (R = −0.152; p = 0.25), thus degradation was similar across concentrations.
Ice packs: Mean baseline ACTH concentration for plasma stored between ice packs at −20°C was 86 pmol/L, and declined 1.1% to 85 pmol/L by day 7 (Table 1). Degradation of ACTH was not observed by day 7 in either the ACTH concentrations (p > 0.36) or the %CFB (p > 0.24). The ACTH concentrations in plasma stored between ice packs at −20°C were 10–265 pmol/L on day 0 and 10–261 pmol/L at 7 d. Correlations between ACTH concentration and %CFB were not observed (R = 0.092; p = 0.60), thus degradation was similar across concentrations.
Storing ACTH samples resulted in conservation of plasma ACTH, with no difference from baseline for up to 60 d at −80°C and up to 30 d at −20°C. Degradation of plasma ACTH was observed on day 90 in −80°C stored samples and on day 60 at −20°C. Degradation was not observed in ACTH from plasma samples stored between ice packs at −20°C for 7 d (Fig. 1).
Adrenocorticotropic hormone (ACTH) concentration (pmol/L) of 12 equine plasma samples stored under different conditions for up to 90 d. Error bars indicate standard error within the horse. Compared to baseline samples, freezing plasma at −20°C or at −80°C resulted in degradation of plasma ACTH at 60 and 90 d, respectively. Bars marked with asterisk (*) denote a significant difference from day 0 (p = 0.03). There was no degradation after 7 d when stored between ice packs at −20°C.
Footnotes
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with repect to the research, authorship, and/or publication of this article: D Neal and S Grubbs are employed by Boehringer Ingelheim Vetmedica, but this sponsor had no influence on study design or interpretation.
Funding
This work was funded through Boehringer Ingelheim Vetmedica (Duluth, GA) and The John C. Miller Chair of Equine Reproduction at Middle Tennessee State University (Murfreesboro, TN).
