Stability and validity of intact parathyroid hormone levels in different sample types and storage conditions

Abstract Background Several pre‐analytical factors can affect the measurement of intact Parathyroid Hormone (IPTH). In this study, we have investigated the effects of using different types of tubes, time elapsed before separation, and storage conditions over time on the measured values of IPTH. Method Blood samples from 30 subjects were collected into plain, SST, and EDTA tubes. All serum and plasma were separated immediately (first set) and after 2 hrs delay (second set). The first set of samples were aliquoted and stored at RT (25°C), at fridge (4°C), and freezer (−20°C). IPTH was measured in all the stored aliquots at 2,4, and 8 days after collection using Architect analyzer. Results Paired T test and ANOVA repeated measures showed no significant difference between IPTH levels in all tubes. The second set of serum and plasma were significantly lower (3.8% and 7.4%, p < 0.001, respectively) when compared to samples measured initially. Serum samples stored at RT were significantly lower (by 45%,59%, and 77%) on days 2,4, and 8 when compared to the initial time (p < 0.001 in all cases). Plasma samples stored at RT, were significantly lower on day 8 after collection, by 30.8% (p < 0.001). These differences would be clinically important. Conclusion Plasma IPTH can be stored at RT for up to four days. Both plasma and serum IPTH are not affected by a delay in the separation of up to two h and they can be stored for up to 8 days in a fridge or freezer without any clinically significant changes in their values.


| INTRODUC TI ON
The parathyroid glands are responsible for maintaining blood calcium level; this occurs by their secretion of parathyroid hormone (PTH) which exerts its effect on bone, kidney, and gastrointestinal system to maintain calcium level. 1 Whilst many blood analytes are quite stable, and their measurements remain unaltered in different conditions, others are sensitive to several factors including time of sample collection, tube type, sample transportation, and sample storage conditions. Parathyroid hormone is critically important in clinical diagnosis of serum calcium abnormalities, and it has been shown that measured PTH levels vary widely between different commercial kits due to the lack of PTH standardization. 2 The optimum procedure to measure IPTH is rapid centrifugation after blood collection, followed by immediate measurement. This cannot always be guaranteed in many laboratories, however, due to factors such as test availability, equipment readiness, batching analysis, and distance from clinics to the laboratory, etc. Such delays could affect the stability of the IPTH, leading to measurement errors that may, in turn, affect the treatment of patients.
It has shown that about 70% of the laboratory errors are preanalytical. 3 The issue of pre-analytical stability has been widely investigated in different biological analytes in order to establish the appropriate procedures to give reliable results. Therefore, a number of studies have sought to identify the factors affecting the pre-analytical stability of IPTH in order to help develop more appropriate procedures but these have had inconsistent outcomes, are outdated, and/or contradict each other in respect to optimum storage conditions (see the summary in Table 1). Moreover, previous studies have tended to focus on individual factors, such as sample type (serum vs. plasma), 4,5 storage temperature, 6,7 and storage time. 8 No study has attempted to evaluate all the major factors in the same protocol.
The aim of this study, therefore, is to investigate the following major factors contributing to variations in measured IPTH levels: the use of different types of blood collection tubes, the time elapsed before separation, and storage conditions over time.

| Study design
There were 30 participants in total: 20 male and 10 female healthyadult subjects. The mean ± SD of their age was 33.7 ± 9.0 years and BMI 26.2 ± 2.2 kg/m 2 , and of their systolic and diastolic blood pressure, 124 ± 14 mmHg and 73 ± 10 mmHg, respectively. In order to maintain consistent pre-analytical conditions between different samples, blood samples were collected into five tubes (

| Collection of blood specimens
Peripheral venous blood samples were collected as recommended by the Clinical Laboratory Standards Institute (CLSI), Document GP41. 9 In brief, the samples were divided into two groups. The first group (first set) comprised three of the tubes taken from each participant (one plain, one plasma EDTA, and one SST). For these samples, serum and plasma were separated immediately after collection by centrifugation for 10 min at 3500 rpm. An aliquot of separated serum and plasma from each tube was immediately analyzed for IPTH according to the method described below. These provided the initial time (baseline) data. The remaining separated serum and plasma were aliquoted into plastic tubes and stored at different conditions: room temperature (24°C), fridge (4°C), and freezer (−20°C). These were subsequently analyzed at day 2, day 4, and day 8.
The second group (second set) comprised two of the tubes taken from each participant (one plain and one plasma EDTA). For these samples, separation was delayed for two h after collection. At that point, the same separation protocol was applied as for the first set, immediately followed by analysis for IPTH. The means of the IPTH content of the serum and plasma samples were compared. In addition, the means of the IPTH content of the delayed serum and plasma were compared to the initial time data, as were the means of the stored serum and plasma from the first set, in order to assess possible deterioration in stability.
To confirm that our patients were healthy, other IPTH related tests were performed using the initial blood samples. They were as follows (mean ± SD): calcium (2.46 ± 0.07 mmol/L), phosphate

| Laboratory Testing
The IPTH assay was done using an Architect i2000 instrument (Abbott Diagnostics). The assay is based on chemiluminescent microparticle immunoassay (CMIA) for quantitative determination of IPTH in human serum and plasma. The kit insert instructions were followed to ensure the reliability of the results. Reagents were donated by the Abbott Company (Medi-Serve). The Architect IPTH assay was investigated for within and between run precision (percentage of coefficient of variation (%CV)). Between

| Statistical analysis
Data were collected for both anthropometric and testing parameters using data entry sheets. The Statistical Package for Social Sciences (SPSS), version 25 (IBM) was utilized for data management.
A test for normality was performed on all continuous variables, and was normal, as evidenced by a non-significant Shapiro-Wilk test.
Accordingly, parametric statistical tests were adopted for comparisons. One-way ANOVA, paired T test, ANOVA with repeated measures, Least significant difference, and percent change tests were applied, and data were reported as mean ± SD and statistical significance was determined at p < 0.05.

| Comparison between different sample types
When plasma and serum were separated immediately after collection, the recorded IPTH mean values ± SD in the three tube types were as follows: in plasma EDTA tubes (67.3 ± 26.0 pg/mL), in SST serum tubes (65.9 ± 24.8 pg/mL), and in plain serum tubes (67.8 ± 25.9 pg/mL). One-way ANOVA showed that there was no statistically significant for the IPTH values using these different types of tubes at the p < 0.05 level F (2,87) = 0.047, p = 0.954.

| Total allowable error (TEa) and Total observed error (TEo):
Data on total changes of analyte concentrations have important clinical implications in patient management. One of the most widely accepted approaches for this purpose is to compare the percentage of change (TEo) with the quality parameter called the total allowable error (TEa). 11 TEa is a quality requirement that sets a limit for the summation of imprecision (random error) and bias (inaccuracy, or systematic error) that are tolerable in a single test result to ensure clinical usefulness. TEa can be defined as the error that is allowable when compared to the reference or original value without affecting the interpretation of the test result or exceeding the clinical decision thresholds. It can be expressed as the maximum error and imprecision which can be allowed for each analyte. Figure 3 The TEa for IPTH is 30% as recently recommended by the Clinical Laboratory Improvement Amendments. 12 This means that observed total error (TEo) values of more than ± 30% of the reference value will be clinically significant. Thus, TEo values not exceeding the TEa have no clinical significance on patient outcomes. In this study, the IPTH value at the initial time in each condition will be taken as the reference value.
The results in Table 2

| DISCUSS ION
Laboratory test results contribute to about 70% of all clinical decisions. 13 Therefore, for reliable clinical decision, it is important to investigate different laboratory tests at different conditions.
Parathyroid hormone (PTH) is the main regulator for calcium and phosphorous levels in blood circulation through its direct effects on the cells of different vital organs. 14 There are many factors that can affect the reliability of PTH results. 15,16 In this study, we provide information about IPTH pre-analytical stability that should be useful.
We controlled subject variability by collecting samples from the same subject in all tube types at the same time. To control instrument variability, meanwhile, we analyzed the initial time samples immediately on collection using the same instrument. In practice, it is useful to have some freedom in the choice of tube type. When many tests need to be done simultaneously on a sample at the same laboratory using just a single tube helps minimize the volume of blood taken from patients, controls costs and supports the quality of results.
Our results showed that there were no statistically significant differences in IPTH levels between samples collected using plain, SST, or EDTA tubes. Interestingly, a previous study to assess IPTH levels in serum and EDTA plasma samples did indicate a statistically significant difference when an Immulite 2000 analyzer was used for analysis. 7 That study also showed that these differences were not apparent when an Architect analyzer was used, and thus may be due to the different reagent components utilized in the two devices. In addition to the study of La'ulu and Roberts included fewer subjects than ours. Another implication of our finding is that there were no statistically significant differences between tubes containing a gel separator in the SST had no significant effect on IPTH results when compared to serum or EDTA plasma.
Overall, our findings in respect to the impact of the choice of collection tube support the use of the smallest number of tubes for blood collection that would be necessary for many clinical tests. This will allow clinical laboratories to minimize the volume of blood taken from patients, and reduce costs. Although in this study we did not investigate the desirable quality specifications components but they can be estimated based on within and between individuals variations as described by Fraser and Harris. 11 The desirable specifications were summarized from previous studies by Ricos. 18 The IPTH components were not included in the listed summary of Ricos but they were recently published. 19 Our study does have some limitations, however. Firstly, it is limited to a specific type of instrument, the Architect i2000.
Although most analyzers have the same principle of analysis, they use different reagent antibodies and accordingly operate on different specifications for IPTH detection. In addition, although the Becton Dickenson collection tubes utilized in our study are widely used, tubes from different companies lead to different outcomes.
The final limitation of our study was that it was conducted on healthy individuals only rather than on patients with pathological conditions or dialysis.
In conclusion, we show that IPTH is more stable than previous research suggests. IPTH can be tested using plasma or serum samples with and without gel. A delay between sample separation and analysis is acceptable, but the most accurate results are obtained if IPTH is analyzed immediately after blood collection. If this is not possible, storage in a fridge or freezer would be recommended (for up to 8 days).
In general, plasma IPTH has more stability than serum and can be stored at room temperature for up to 8 days.

ACK N O LED G EM ENTS
We are grateful for the kind financial support and collaboration re-

CO N FLI C T I NTE R E S T
The authors declare no competing interests.

AUTH O R CO NTR I B UTI O N S
HK researched the literature and conceived the study. Recruitment of subjects and analytical work was carried out by HK, HF, and MD.

D ECL A R ATI O N S
The authors have no declarations.

E TH I C A L A PPROVA L
Ethical approval for this study was approved by the institutional re-

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.