Abstract
Background
The stability of biochemical analytes has already been investigated, but results strongly differ depending on parameters, methodologies, and sample storage times. We investigated the stability for many biochemical parameters after different storage times of both whole blood and plasma, in order to define acceptable pre‐ and postcentrifugation delays in hospital laboratories.
Methods
Twenty‐four analytes were measured (Modular® Roche analyzer) in plasma obtained from blood collected into lithium heparin gel tubes, after 2–6 hr of storage at room temperature either before (n = 28: stability in whole blood) or after (n = 21: stability in plasma) centrifugation. Variations in concentrations were expressed as mean bias from baseline, using the analytical change limit (ACL%) or the reference change value (RCV%) as acceptance limit.
Results
In tubes stored before centrifugation, mean plasma concentrations significantly decreased after 3 hr for phosphorus (–6.1% [95% CI: –7.4 to –4.7%]; ACL 4.62%) and lactate dehydrogenase (LDH; –5.7% [95% CI: –7.4 to –4.1%]; ACL 5.17%), and slightly decreased after 6 hr for potassium (–2.9% [95% CI: –5.3 to –0.5%]; ACL 4.13%). In plasma stored after centrifugation, mean concentrations decreased after 6 hr for bicarbonates (–19.7% [95% CI: –22.9 to –16.5%]; ACL 15.4%), and moderately increased after 4 hr for LDH (+6.0% [95% CI: +4.3 to +7.6%]; ACL 5.17%). Based on RCV, all the analytes can be considered stable up to 6 hr, whether before or after centrifugation.
Conclusion
This study proposes acceptable delays for most biochemical tests on lithium heparin gel tubes arriving at the laboratory or needing to be reanalyzed.
Keywords: stability, storage time, preanalytical, biochemical analytes, whole blood, plasma
INTRODUCTION
In France, the current constrained economic context leads the major hospitals to group their biological activities in centralized automated platforms, but blood is collected in centers that are located in the periphery, sometimes very far from the central laboratory. Such organization may lengthen the transport times of blood samples, especially in big cities where the traffic is heavy at peak hours. Whether or not blood collection centers or clinical services are equipped with centrifuges, whole blood and plasma stabilities of prescribed analytes become strongly dependent on transport times, and thus must be known and controlled by the laboratory staff 1. Indeed, a prolonged contact between plasma or serum and blood cells leads to significant variations of concentrations for some biochemical analytes, as shown for potassium (K), lactate dehydrogenase (LDH), or phosphorus (PHOS), for example 2, 3, 4. In addition to storage time, the temperature of storage also influences the stability of some analytes 5, 6, 7.
Many studies have investigated the stability of routine biochemical analytes. Nevertheless, results and conclusions often differ substantially depending on the studied variables: different panels of analytes, measured in different matrixes (whole blood, serum, plasma with different anticoagulants, blood collected in tubes with or without separator gel), evaluation on plasma/serum pools or from primary blood tubes, collected from different populations (healthy subjects or hospitalized patients), with variable group sizes, different storage times (single time or kinetics, for hours and/or days), different temperature conditions (unspecified or controlled room temperature [RT], –20°C, –80°C), different technologies and analyzers, different acceptance limits (analytical and/or clinical change limits) or statistical models (ANOVA, paired t‐test or Wilcoxon rank test), different result expressions (absolute or relative biases, means or medians, with or without standard deviations, 95% confidence intervals [CIs] or minimum/maximum ranges).
Given all these differences regarding study designs, methodologies and results, we aimed to evaluate the stability for 24 biochemical analytes according to a pragmatic approach adapted to our routine lab practice: the blood was collected into lithium heparin gel tubes, stored 2–6 hr at RT before (stability in whole blood) or after (stability in plasma) centrifugation, followed by measures of all the analytes.
MATERIALS AND METHODS
Whole blood and plasma stabilities were assessed using venous blood collected in lithium heparin tubes (18 U/ml lithium heparin, separator gel, ref #474080, Greiner Bio‐One, Austria), gently inverted eight times after sampling. Centrifugation conditions were 1,885 g for 10 min at +17°C using a Rotina 380‐R® centrifuge (Hettich, Germany), with the aim to reproduce the conditions of our Modular Preanalytics (Roche Diagnostics, Germany). The following 24 biochemical analytes were quantified on Modular P800® analyzer (Roche Diagnostics): albumin (ALB, by colorimetry), alkaline phosphatase (ALP), alanine aminotransferase (ALT, with pyridoxal 5′‐phosphate), amylase (AMY), aspartate aminotransferase (AST, with pyridoxal 5′‐phosphate), total bilirubin (BILT), calcium (Ca), total cholesterol (CHOL), creatine kinase (CK), chloride (Cl), total carbon dioxide (CO2), creatinine (CREA), γ‐glutamyltransferase (GGT), haptoglobin (HPT), K, LDH, lipase (LIP), magnesium (Mg), sodium (Na), phosphorus (PHOS), triglycerides (TG), total protein (TP), uric acid (UA), and urea (UR). All the principles of methods, technologies, wavelengths, and reagent references are provided in Supporting Information Table A.
Stability in Whole Blood
Five heparinized blood collection tubes were collected, fully filled, from 28 healthy volunteers in sitting position with a total collection time of 5 min at an average rate of five volunteers per day. The first tube was kept in upright position for exactly 2 hr (baseline time) at RT (mean 21.3 ± 1.8°C), centrifuged, and then all the plasma parameters were measured. The second, third, fourth, and fifth tubes were treated following the same procedure, but after being kept in upright position during 3, 4, 5, and 6 hr, respectively.
Stability in Plasma
Maximally filled heparin tubes from 21 hospitalized patients were selected. They were received in our laboratory with a mean delay of 2 hr ± 18 min (baseline time). These tubes were immediately centrifuged and measurements were made on analyzer. Then these tubes were kept at rest, in upright position, at RT, and reassayed for all measurements after 2, 4, and 6 hr of storage. Each time, all tubes were immediately recapped at the output of the analyzer.
CALCULATIONS AND STATISTICS
For each analyte and each storage time, mean %change of concentration from baseline was determined using the formula: ±mean bias% = 100 × (measured value − baseline value)/baseline value, wherein baseline value was the plasma concentration for the first tube (kept ∼2 hr) measured immediately after centrifugation. Mean biases were given with their 95% CI, and were considered significant if the absolute value of the bias was above the analytical change limit (ACL), that is, the maximum analytical variation; ACL is equal to 1.96 × √2 × CVa 8, wherein 1.96 × √2 corresponds to the standard deviation for the bidirectional probability of change fixed at 95%, and CVa to the inter‐assay imprecision calculated from Roche Diagnostics or Biorad Quality Control values collected over an 11‐month period and chosen with the closest concentrations from the mean baseline values. In addition, biases were also interpreted regarding the analytical plus within‐subject biological variations (CVw, available from Westgard website 9) corresponding to the reference change value (RCV = 1.96 × √2 × √(CVa2 + CVw2)). Calculated CVa, CVw, ACL, and RCV are described in Supporting Information Table B.
RESULTS
In whole blood study (Table 1), considering the ACL, mean concentrations decreased significantly after 4 hr for LDH (–5.7% [95% CI: –7.4 to –4.1%]; ACL = 5.2; Fig. 1A) and PHOS (–6.1% [95% CI: –7.4 to –4.7%]; ACL = 4.6; Fig. 1B), but not significantly for K, even after 6 hr (–2.9% [95% CI: –5.3 to –0.5%]; ACL = 4.1; Fig. 1C). In heparinized plasma study (Table 2), mean concentrations can be considered stable up to 4 hr for bicarbonates (–12.4% [95% CI: –14.5 to –10.3%]; ACL = 15.4; Fig. 2A), but then they decreased significantly, whereas mean concentrations were stable up to 2 hr for LDH and then increased significantly after 4 hr (+6.0% [95% CI: +4.3 to +7.6%]; ACL = 5.2; Fig. 2B). Based on ACL, all the other analytes were considered stable on whole blood and plasma up to 6 hr. Regarding the RCV, 24 analytes were considered stable for 6 hr in both substudies.
Table 1.
Stability of Biochemical Tests on Plasma Obtained From Blood Collected on Lithium Heparin Gel Tubes Stored up to 6 hr Before Centrifugation
| Mean bias (%; 95% CI) | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Analyte | Unit | n | Mean baseline value (range)a | +3 hr | +4 hr | +5 hr | +6 hr | ACL (%)b | Biasc | Acceptable delay (hr)d | |||||
| ALB | g/l | 28 | 47 | (37−52) | −0.22 | (−0.93; +0.48) | −0.44 | (−1.35; +0.46) | +0.28 | (−0.68; +1.25) | +0.09 | (−0.82; +1.00) | 7.32 | ns | +6 hr |
| ALP | U/l | 28 | 60 | (32−110) | −0.86 | (−1.56; −0.17) | −1.97 | (−2.69; −1.26) | −2.78 | (−3.66; −1.90) | −2.36 | (−3.05; −1.66) | 9.11 | ns | +6 hr |
| ALT | U/l | 28 | 21 | (6−82) | −0.85 | (−6.37; +4.67) | −1.41 | (−5.97; +3.15) | −0.40 | (−3.70; +2.89) | −1.99 | (−6.98; +3.01) | 10.0 | ns | +6 hr |
| AMY | U/l | 28 | 72 | (39−118) | +0.29 | (−0.37; +0.95) | +0.01 | (−0.92; +0.94) | −0.47 | (−1.62; +0.68) | −0.24 | (−1.21; +0.72) | 5.38 | ns | +6 hr |
| AST | U/l | 28 | 22 | (15−39) | −0.66 | (−4.16; +2.83) | −0.48 | (−4.19; +3.22) | +0.06 | (−3.37; +3.48) | +0.12 | (−2.94; +3.19) | 9.89 | ns | +6 hr |
| BILT | μmol/l | 28 | 10 | (3−33) | +2.48 | (−0.52; +5.48) | +1.40 | (−1.27; +4.06) | +3.23 | (+0.11; +6.35) | +3.18 | (−0.36; +6.72) | 8.71 | ns | +6 hr |
| CA | mmol/l | 28 | 2.39 | (2.15−2.80) | −0.21 | (−1.55; +1.14) | +0.21 | (−1.40; +1.83) | +0.68 | (−0.82; +2.19) | +0.03 | (−1.66; +1.72) | 5.86 | ns | +6 hr |
| CHOL | mmol/l | 28 | 5.20 | (3.43−7.25) | +0.41 | (−0.72; +1.55) | +0.33 | (−1.10; +1.76) | +0.13 | (−1.28; +1.55) | +0.29 | (−1.01; +1.59) | 5.45 | ns | +6 hr |
| CK | U/l | 28 | 125 | (49−358) | −0.48 | (−1.74; +0.78) | −0.93 | (−2.05; +0.18) | −1.30 | (−2.55; −0.05) | −0.51 | (−1.51; +0.48) | 5.57 | ns | +6 hr |
| Cl | mmol/l | 28 | 101 | (98−104) | −0.03 | (−0.47; +0.40) | −0.07 | (−0.41; +0.28) | −0.10 | (−0.52; +0.33) | −0.28 | (−0.80; +0.24) | 3.77 | ns | +6 hr |
| CO2 | mmol/l | 28 | 25.9 | (22.6−32.6) | −0.85 | (−1.82; +0.11) | −0.60 | (−1.49; +0.29) | −1.97 | (−3.06; −0.88) | −2.48 | (−3.82; −1.14) | 15.4 | ns | +6 hr |
| CREA | μmol/l | 28 | 72 | (45−104) | −0.47 | (−2.52; +1.58) | +0.65 | (−1.47; +2.77) | −0.77 | (−3.02; +1.48) | −0.46 | (−2.50; +1.58) | 8.83 | ns | +6 hr |
| GGT | U/l | 28 | 23 | (10−54) | −0.73 | (−4.94; +3.47) | −1.92 | (−5.33; +1.49) | −2.59 | (−5.49; +0.31) | +0.57 | (−1.81; +2.95) | 8.12 | ns | +6 hr |
| HPT | g/l | 28 | 1.07 | (0.38−1.60) | −0.48 | (−1.43; +0.47) | −0.84 | (−1.92; +0.24) | −1.31 | (−2.32; −0.29) | −0.54 | (−1.62; +0.53) | 5.90 | ns | +6 hr |
| K | mmol/l | 28 | 4.0 | (3.4−4.7) | −0.95 | (−2.10; +0.21) | −2.33 | (−3.65; −1.01) | −2.63 | (−3.94; −1.33) | −2.89 | (−5.31; −0.47) | 4.13 | ↓ | +5 to +6 hr |
| LDH | U/l | 28 | 345 | (227−481) | −2.39 | (−4.23; −0.55) | −5.73 | (−7.38; −4.07) | −4.91 | (−6.71; −3.10) | −3.74 | (−5.63; −1.86) | 5.17 | ↓ | +3 hr |
| LIP | U/l | 28 | 39 | (20−114) | +0.30 | (−0.92; +1.51) | −0.20 | (−1.51; +1.10) | −0.25 | (−1.59; +1.09) | +1.04 | (+0.11; +1.97) | 7.86 | ns | +6 hr |
| Mg | mmol/l | 26 | 0.87 | (0.75−1.03) | −0.47 | (−1.28; +0.35) | +0.93 | (−0.17; +2.02) | +1.84 | (−0.33; +4.01) | +1.61 | (+0.15; +3.06) | 5.36 | ns | +6 hr |
| Na | mmol/l | 28 | 141 | (138−145) | +0.00 | (−0.27; +0.27) | −0.17 | (−0.45; +0.10) | +0.03 | (−0.21; +0.27) | +0.10 | (−0.23; +0.44) | 3.20 | ns | +6 hr |
| PHOS | mmol/l | 28 | 1.03 | (0.55−1.35) | −3.36 | (−4.29; −2.43) | −6.05 | (−7.40; −4.70) | −8.61 | (−9.85; −7.36) | −9.85 | (−11.7; −7.98) | 4.62 | ↓ | +3 hr |
| TG | mmol/l | 28 | 0.92 | (0.38−2.15) | −0.18 | (−1.40; +1.04) | −1.05 | (−2.19; +0.08) | −2.40 | (−3.48; −1.33) | −1.68 | (−3.23; −0.14) | 7.84 | ns | +6 hr |
| TP | g/l | 28 | 75 | (66−81) | +0.20 | (−0.27; +0.67) | −0.14 | (−0.81; +0.53) | −0.49 | (−1.39; +0.41) | +0.01 | (−0.47; +0.50) | 4.59 | ns | +6 hr |
| UA | μmol/l | 28 | 250 | (144−390) | −0.49 | (−0.95; −0.02) | −0.53 | (−0.98; −0.08) | −0.72 | (−1.06; −0.37) | −0.92 | (−1.31; −0.52) | 3.84 | ns | +6 hr |
| UR | mmol/l | 28 | 4.8 | (2.5−8.7) | +1.10 | (−0.46; +2.65) | +0.11 | (−1.95; +2.18) | +1.92 | (+0.45; +3.40) | +2.26 | (+0.56; +3.97) | 7.61 | ns | +6 hr |
Mean heparinized plasma concentration measured after a 2‐hr storage of whole blood, considered the baseline value.
ACL, analytical change limit = 1.96 × √2 × CVa.
Bias considered significant if absolute value greater than ACL in an increasing (↑) or decreasing (↓) change, or without significant variation (ns).
The boldfaced values correspond to the acceptable delay (hr) for which the period of stability was significantly shortened with respect to the total duration of the test.
Figure 1.
The solid black curves represent the mean biases from baseline for LDH (A), PHOS (B), and K (C) plasma concentrations measured on lithium heparin gel tubes stored at room temperature up to 6 hr before centrifugation (n = 28). The two bordered long dashed curves on either side represent the 95% confidence interval of mean curve, and the short dashed lines correspond to the analytical change limit equal to 1.96 × √2 × CVa. The gray curves correspond to the biases from baseline for each subject. ACL, analytical change limit; K, potassium; LDH, lactate dehydrogenase; PHOS, phosphorus.
Table 2.
Stability of Biochemical Tests on Heparinized Plasma Stored up to 6 hr After Centrifugation
| Mean bias (%; 95% CI) | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Analyte | Unit | n | Mean baseline value (range)a | +2 hr | +4 hr | +6 hr | ACL (%)b | Biasc | Acceptable delay (hr)d | ||||
| ALB | g/l | 21 | 34 | (20–47) | +1.21 | (–0.24; +2.67) | +1.93 | (+0.92; +2.94) | +2.24 | (+1.32; +3.17) | 7.32 | ns | +6 hr |
| ALP | U/l | 21 | 111 | (34–357) | –0.41 | (–1.51; +0.69) | –0.60 | (–1.52; +0.32) | +1.02 | (–0.08; +2.12) | 9.11 | ns | +6 hr |
| ALT | U/l | 21 | 43 | (7–126) | +2.64 | (+0.35; +4.94) | +3.28 | (+0.05; +6.50) | +2.22 | (+0.21; +4.22) | 10.0 | ns | +6 hr |
| AMY | U/l | 21 | 68 | (17–152) | +0.47 | (–0.55; +1.49) | +0.47 | (–0.52; +1.46) | +0.52 | (–2.63; +3.67) | 5.38 | ns | +6 hr |
| AST | U/l | 21 | 31 | (12–85) | +2.11 | (–0.14; +4.35) | +4.40 | (+2.19; +6.62) | +6.04 | (+3.17; +8.91) | 9.89 | ns | +6 hr |
| BILT | μmol/l | 21 | 20 | (3–103) | –1.53 | (–5.39; +2.32) | –2.31 | (–7.53; +2.92) | +0.09 | (–1.27; +1.46) | 8.71 | ns | +6 hr |
| CA | mmol/l | 21 | 2.23 | (1.85–2.58) | +0.52 | (–0.27; +1.30) | +1.38 | (+0.76; +2.00) | +1.71 | (+0.97; +2.45) | 5.86 | ns | +6 hr |
| CHOL | mmol/l | 21 | 4.06 | (1.83–7.04) | +1.37 | (+0.36; +2.39) | +1.28 | (+0.20; +2.35) | +1.87 | (+0.88; +2.86) | 5.45 | ns | +6 hr |
| CK | U/l | 21 | 82 | (17–218) | –1.23 | (–3.27; +0.81) | +0.11 | (–2.71; +2.92) | +1.41 | (–0.39; +3.21) | 5.57 | ns | +6 hr |
| Cl | mmol/l | 21 | 101 | (84–108) | +0.36 | (–0.77; +1.49) | +0.59 | (–0.42; +1.61) | +0.41 | (–0.63; +1.45) | 3.77 | ns | +6 hr |
| CO2 | mmol/l | 21 | 23.8 | (19.5–30.7) | –5.50 | (–7.10; –3.91) | –12.4 | (–14.5; –10.3) | –19.7 | (–22.9; –16.5) | 15.4 | ↓ | +4 hr |
| CREA | μmol/l | 21 | 115 | (33–474) | –0.21 | (–2.50; +2.08) | –0.03 | (–1.46; +1.39) | +5.16 | (+2.48; +7.83) | 8.83 | ns | +6 hr |
| GGT | U/l | 21 | 126 | (15–488) | –0.87 | (–2.05; +0.32) | +0.40 | (–1.36; +2.15) | +2.16 | (+0.53; +3.80) | 8.12 | ns | +6 hr |
| HPT | g/l | 21 | 1.66 | (0.44–3.29) | +1.59 | (+0.88; +2.31) | +1.81 | (+1.07; +2.54) | +2.71 | (+1.58; +3.84) | 5.90 | ns | +6 hr |
| K | mmol/l | 21 | 4.0 | (3.2–4.9) | +0.17 | (–1.20; +1.54) | +1.14 | (–0.39; +2.67) | +1.88 | (+0.36; +3.41) | 4.13 | ns | +6 hr |
| LDH | U/l | 21 | 490 | (221–1027) | +3.04 | (+2.14; +3.94) | +5.97 | (+4.31; +7.64) | +9.67 | (+7.13; +12.2) | 5.17 | ↑ | +2 hr |
| LIP | U/l | 21 | 49 | (9–113) | –1.64 | (–2.78; –0.50) | –3.29 | (–5.12; –1.45) | –0.60 | (–1.50; +0.29) | 7.86 | ns | +6 hr |
| Mg | mmol/l | 21 | 0.83 | (0.64–1.14) | +1.43 | (+0.18; +2.69) | +2.36 | (+1.31; +3.41) | +2.03 | (+1.01; +3.05) | 5.36 | ns | +6 hr |
| Na | mmol/l | 21 | 138 | (115–149) | +0.64 | (–0.40; +1.68) | +1.03 | (+0.11; +1.96) | +1.42 | (+0.45; +2.39) | 3.20 | ns | +6 hr |
| PHOS | mmol/l | 21 | 1.03 | (0.33–1.84) | +0.37 | (–0.27; +1.01) | +0.96 | (+0.02; +1.90) | +2.70 | (+1.34; +4.05) | 4.62 | ns | +6 hr |
| TG | mmol/l | 21 | 1.31 | (0.53–2.73) | +0.41 | (–0.70; +1.51) | +0.30 | (–0.83; +1.42) | –0.01 | (–1.41; +1.39) | 7.84 | ns | +6 hr |
| TP | g/l | 21 | 62 | (42–83) | +0.21 | (–0.44; +0.86) | +0.91 | (+0.38; +1.43) | +1.40 | (+0.80; +1.99) | 4.59 | ns | +6 hr |
| UA | μmol/l | 21 | 270 | (67–708) | +0.82 | (+0.43; +1.20) | +1.40 | (+0.88; +1.91) | +2.24 | (+1.56; +2.93) | 3.84 | ns | +6 hr |
| UR | mmol/l | 21 | 10.0 | (3.8–33.7) | +1.49 | (–0.07; +3.06) | +1.91 | (+0.36; +3.47) | +2.16 | (+0.70; +3.62) | 7.61 | ns | +6 hr |
Mean heparinized plasma concentration measured 2 hr ± 18 min following blood sampling, considered the baseline value.
ACL, analytical change limit = 1.96 × √2 × CVa.
Bias considered significant if absolute value greater than ACL in an increasing (↑) or decreasing (↓) change, or without significant variation (ns).
The boldfaced values correspond to the acceptable delay (hr) for which the period of stability was significantly shortened with respect to the total duration of the test.
Figure 2.
The solid black curves represent the mean biases from baseline for CO2 (A) and LDH (B) concentrations measured on lithium heparin plasma stored at room temperature up to 6 hr after centrifugation (n = 21). The two bordered long dashed curves on either side represent the 95% confidence interval of mean curve, and the short dashed lines correspond to the analytical change limit equal to 1.96 × √2 × CVa. The gray curves correspond to the biases from baseline for each subject. ACL, analytical change limit; CO2, total carbon dioxide; LDH, lactate dehydrogenase.
DISCUSSION
We evaluated the influence of storage time, in kinetics up to 6 hr and at RT, on 24 plasma routine biochemical analytes measured on lithium heparin gel tubes stored before (stability in whole blood) and after (stability in plasma) centrifugation.
In both substudies, we selected the baseline time at 2 hr after blood sampling because 2 hr corresponds to the average delivery time of heparin tubes at our lab from faraway sites, by pneumatic transport or directly brought by the medical staff (expressed as cumulative frequency, about 52% of tubes arrive within 2 hr, 80% within 3 hr, and 95% within 4 hr; data not shown). In the same manner, Doumas et al. retained tubes centrifuged within 2 hr after collection, and used this 2‐hr delay as baseline time 10. Furthermore, many electrolytes, including K and PHOS, have been shown to be stable within plasma in the first 2 hr after blood sampling 11.
Stability in Whole Blood
We have shown that mean PHOS concentration decreased significantly after 4 hr (‒6.1%) of contact between plasma and blood cells. Such decrease had already been shown in previous studies, after 4–6 hr (‒4.1% to ‒6.4%) at 20°C 12, after 2–4 hr (‒4.4% to ‒7.9%) at 25°C 5, and after 10 hr (‒11.2%) of storage at 21°C 2. Another study using ANOVA as significant differences showed that PHOS concentration decreased after 8 hr at 25°C due to glycolysis, or after 16 hr using significant change limit, then progressively increased after 32–56 hr of contact 6 due to hydrolysis of intraerythrocytic phosphate esters and leakage of the inorganic part 13. The decrease of PHOS in whole blood disappeared when tubes were stored at 4°C, at least up to 4 hr 5. In our study, LDH concentrations slightly decreased (‒5.7% after 4 hr), as well as K concentrations after 6 hr (maximum mean bias ‒2.9%). Stahl et al. found negligible variations of stability in whole blood at 23°C for LDH (–3.3%, –5.5%, –1.3% at 2, 4, and 6 hr, respectively) and for K (–2.6%, –0.5%, –3.1% at 2, 4, and 6 hr, respectively) 12. Boyanton and Blick did not show any significant variation of stability for LDH and K after 4 and 8 hr, except a slight decrease in K concentration (–0.1 mmol/l) at 4 hr 6. Nonetheless, one must remember that stability of LDH and K depends on the temperature, which is known to influence enzymatic activities 13. In particular, the Na+,K+‐ATPase activity has been shown to be stimulated at 25°C, leading to a decrease in extracellular K concentration, and to diminish at 17°C, leading to the increase in extracellular K 13, as elsewhere observed in plasma after 6 hr of storage at +8°C 14. Given this, the increases in K concentration at 25°C, shown in Oddoze et al., on whole blood after 2 hr (+3.5%) and 4 hr (+4.3%) remain somewhat questionable 5. Considering the RCV 9, the decrease in PHOS concentration after 3–4 hr of whole blood storage and the slight decreases observed for LDH and K concentrations can be considered devoid of clinical significance up to 6 hr (RCV = 23.1%, 24.4%, and 13.4%, respectively; Supporting Information Table B), as previously reported 2. Finally, it must be noted that the design of our substudy on whole blood did not take into account the influence of the pneumatic tube system, which potentially increases the hemolysis degree, thus impacting the concentration of sensitive analytes 15, 16, 17.
Stability in Plasma
We showed that CO2 mean concentration decreases progressively in heparinized plasma, remaining stable up to 4 hr (mean bias –12.4%; ACL 15.4%), and then decreases significantly (mean bias –19.7% at +6 hr). Boyanton and Blick did not show any significant decrease in CO2 in plasma up to 56 hr of storage at 25°C (greatest change –4.8%), neither did Oddoze et al. up to 4 hr (–0.8%) 5, 6. Nevertheless, in accordance with our result, Doumas et al. also observed a growing decrease in plasma bicarbonate stored in capped original tubes (from ‒7% to ‒19% between +4 and +24 hr of storage) 10. Kumar and Karon also showed a significant decrease in bicarbonate concentration (–1 mmol/l) in 20 plasma samples exposed to air for 1 hr, when measured—as we did—using the phosphoenolpyruvate reaction on Roche analyzer 18. Similarly, Kirschbaum showed ∼7% decrease in CO2 concentration in serum exposed to air during 2 hr 19, and Zazra et al. reported a rate loss of CO2 of 2.5 mEq/l/hr from an open autoanalyzer cup 20. Such decreases in CO2 plasma concentrations could be explained, at least in part, by the volatility of dissolved CO2 (∼5% of total CO2), which depends on the time of air exposure and temperature. Moreover, it could be hypothesized that CO2 concentrations also decrease in plasma because of the absence of a major buffer system from red blood cells, mainly the hemoglobin (and thus carbaminohemoglobin ∼10%), as well as carbonic anhydrase and other components allowing the CO2/HCO3 − flux and acid/base equilibrium. Regarding LDH stability in plasma, our results showed an increase in mean concentration of about 6% at +4 hr and 10% at +6 hr. Boyanton and Blick reported a good stability of LDH in plasma over a 56‐hr period (mean 137 U/l with maximum change of 1 U/l) 6, and Heins et al. reported a good stability of LDH in serum stored up to 7 days at RT 7. Oddoze et al., on the other hand, reported an increase in mean concentration of 5.8% after 6 hr in serum kept at 25°C, a variation they considered nonsignificant regarding their acceptance limit of 6.4% 5. In the study of Doumas et al., the mean LDH concentration slightly increased in plasma stored at RT, from 2% to 4% between 4 and 6 hr, but with very wide ranges 10. In accordance, our results showed an increase in the samples of all patients, without exception, but also with wide ranges. We used the pyruvate‐to‐lactate reaction for LDH activity (optimized LDH kit from Roche Diagnostics, Supporting Information Table A), using a first reagent buffered at pH 7.5 including pyruvate and a second one including NADH. Since in this direction, the optimal pH for LDH activity is 7.2 21, the lower the CO2 concentrations, the higher the LDH activity, as suggested by the strong inverse correlation observed on our cumulated data from kinetic (ρs = ‒0.465; P < 0.001, data not shown). Moreover, it is known that blood cells, present in the plasma after centrifugation, are progressively lysed, leading to a release of LDH, K, and PHOS; added to the fact that in the first hours, glycolysis still remains active and consumes glucose, producing pyruvate and lowering pH, associated to an increase in LDH activity 22. Finally, considering the RCV, the maximum variations observed in plasma for CO2 (‒19.7%) and LDH (+9.7%) at 6 hr postcentrifugation can be considered nonsignificant on the clinical level (RCV = 20.3 and 24.4%, respectively).
CONCLUSION
Delivery time before centrifugation leads to a major analytical change in concentrations for PHOS and minor analytical change for LDH and K, whereas storage time after centrifugation impacts total CO2 and LDH. This study provides detailed biases of mean concentrations up to 6 hr of storage time at 21°C, and proposes acceptable delays for common biochemical tests on lithium heparin gel tubes arriving at the laboratory or needing to be reanalyzed.
CONFLICT OF INTEREST
The authors have no conflict of interest to disclose.
Supporting information
Disclaimer: Supplementary materials have been peer‐reviewed but not copyedited.
Supporting Table
ACKNOWLEDGMENTS
The authors thank Paris Descartes Language Center (CdL) for proofreading the manuscript.
REFERENCES
- 1. Dialma P, Piaulenne S, Baty S, Zeitoun T. Preanalytical phase and accreditation: Acceptance criteria for samples of multisite laboratory]. Ann Biol Clin (Paris) 2013;71:121–128. [DOI] [PubMed] [Google Scholar]
- 2. Henriksen LO, Faber NR, Moller MF, Nexo E, Hansen AB. Stability of 35 biochemical and immunological routine tests after 10 hours storage and transport of human whole blood at 21 degrees C. Scand J Clin Lab Invest 2014;74:603–610. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Zhang DJ, Elswick RK, Miller WG, Bailey JL. Effect of serum‐clot contact time on clinical chemistry laboratory results. Clin Chem 1998;44:1325–1333. [PubMed] [Google Scholar]
- 4. Laessig RH, Indriksons AA, Hassemer DJ, Paskey TA, Schwartz TH. Changes in serum chemical values as a result of prolonged contact with the clot. Am J Clin Pathol 1976;66:598–604. [DOI] [PubMed] [Google Scholar]
- 5. Oddoze C, Lombard E, Portugal H. Stability study of 81 analytes in human whole blood, in serum and in plasma. Clin Biochem 2012;45:464–469. [DOI] [PubMed] [Google Scholar]
- 6. Boyanton BL, Jr , Blick KE. Stability studies of twenty‐four analytes in human plasma and serum. Clin Chem 2002;48:2242–2247. [PubMed] [Google Scholar]
- 7. Heins M, Heil W, Withold W. Storage of serum or whole blood samples? Effects of time and temperature on 22 serum analytes. Eur J Clin Chem Clin Biochem 1995;33:231–238. [DOI] [PubMed] [Google Scholar]
- 8. Ricós C, Cava F, Garcia‐Lario JV, et al. The reference change value: A proposal to interpret laboratory reports in serial testing based on biological variation. Scand J Clin Lab Invest 2004;64:175–184. [DOI] [PubMed] [Google Scholar]
- 9. Ricós C, Alvarez V, Cava F, et al. Current databases on biological variation: Pros, cons and progress. Scand J Clin Lab Invest 1999;59:491–500. Available at: http://www.westgard.com/biodatabase1.htm. Accessed in January 2015. [DOI] [PubMed] [Google Scholar]
- 10. Doumas BT, Hause LL, Simuncak DM, Breitenfeld D. Differences between values for plasma and serum in tests performed in the Ektachem 700 XR Analyzer, and evaluation of “plasma separator tubes (PST).” Clin Chem 1989;35:151–153. [PubMed] [Google Scholar]
- 11. Calatayud O, Tenias JM, Castells X. Agreement in the determination of electrolytes in samples of serum and plasma. Stability of these samples kept for 2 hours at room temperature with uncovered tubes. Ann Biol Clin (Paris) 2006;64:157–161. [PubMed] [Google Scholar]
- 12. Stahl M, Brandslund I. Controlled storage conditions prolong stability of biochemical components in whole blood. Clin Chem Lab Med 2005;43:210–215. [DOI] [PubMed] [Google Scholar]
- 13. Ono T, Kitaguchi K, Takehara M, Shiiba M, Hayami K. Serum‐constituents analyses: Effect of duration and temperature of storage of clotted blood. Clin Chem 1981;27:35–38. [PubMed] [Google Scholar]
- 14. Leino A, Koivula MK. Stability of chemical and immunochemical analytes in uncentrifuged plasma samples. Ann Clin Biochem 2009;46:159–161. [DOI] [PubMed] [Google Scholar]
- 15. Böckel‐Frohnhöfer N, Hübner U, Hummel B, Geisel J. Pneumatic tube‐transported blood samples in lithium heparinate gel separator tubes may be more susceptible to haemolysis than blood samples in serum tubes. Scand J Clin Lab Invest 2014;74:599–602. [DOI] [PubMed] [Google Scholar]
- 16. Gomez‐Rioja R, Fernandez‐Calle P, Alcaide MJ, et al. Interindividual variability of hemolysis in plasma samples during pneumatic tube system transport. Clin Chem Lab Med 2013;51:e231–e233. [DOI] [PubMed] [Google Scholar]
- 17. Streichert T, Otto B, Schnabel C, et al. Determination of hemolysis thresholds by the use of data loggers in pneumatic tube systems. Clin Chem 2011;57:1390–1397. [DOI] [PubMed] [Google Scholar]
- 18. Kumar V, Karon BS. Comparison of measured and calculated bicarbonate values. Clin Chem 2008;54:1586–1587. [DOI] [PubMed] [Google Scholar]
- 19. Kirschbaum B. Loss of carbon dioxide from serum samples exposed to air. Effect on blood gas parameters and strong ions. Clin Chim Acta 2003;334:241–244. [DOI] [PubMed] [Google Scholar]
- 20. Zazra JJ, Jani CM, Rosenblum S. Are the results of carbon dioxide analysis affected by shipping blood samples? Am J Kidney Dis 2001;37:1105–1106. [DOI] [PubMed] [Google Scholar]
- 21. Buhl SN, Jackson KY. Optimal conditions and comparison of lactate dehydrogenase catalysis of the lactate‐to‐pyruvate and pyruvate‐to‐lactate reactions in human serum at 25, 30, and 37 degrees C. Clin Chem 1978;24:828–831. [PubMed] [Google Scholar]
- 22. Dimeski G, Solano C, Petroff MK, Hynd M. Centrifugation protocols: Tests to determine optimal lithium heparin and citrate plasma sample quality. Ann Clin Biochem 2011;48:218–222. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Disclaimer: Supplementary materials have been peer‐reviewed but not copyedited.
Supporting Table