See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/319612777

Stability of routine biochemical analytes in whole blood and plasma/serum:

Focus on potassium stability from lithium heparin

Article  in  Clinical Chemistry and Laboratory Medicine · January 2017

DOI: 10.1515/cclm-2017-0292

CITATIONS READS

30 3,945

8 authors, including:

Annie Dupuy Vincent Brunot

Centre Hospitalier Universitaire de Montpellier Centre Hospitalier Universitaire de Montpellier
155 PUBLICATIONS   2,014 CITATIONS    32 PUBLICATIONS   502 CITATIONS   

SEE PROFILE SEE PROFILE

Pascal Philibert
Université de Montpellier - CHRU de Nîmes - Institut de Génétique Humaine
119 PUBLICATIONS   3,039 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Disorders of sex deveopment View project

HDF ICU View project

All content following this page was uploaded by Vincent Brunot on 08 February 2018.

The user has requested enhancement of the downloaded file.

 Clin Chem Lab Med 2017; aop

Anne Marie Dupuy, Jean Paul Cristol*, Bruno Vincent, Anne Sophie Bargnoux,
Mickael Mendes, Pascal Philibert, Kadda Klouche and Stéphanie Badiou

Stability of routine biochemical analytes in whole

blood and plasma/serum: focus on potassium
stability from lithium heparin
https://doi.org/10.1515/cclm-2017-0292 LDH. After prompt centrifugation and storage at 4 °C, sta-
Received April 3, 2017; accepted July 24, 2017 bility was greatly increased up to 48 h for most analytes.
Abstract LDH and bicarbonate had the lowest stability after cen-
trifugation; therefore, no reanalysis of these analytes in a
Background: Blood specimens are transported from clini- centrifuged tube can be allowed.
cal departments to the biochemistry laboratory by hos- Conclusions: Knowledge of analyte stability is crucial to
pital courier service, sometimes over long distances. The interpret biological analysis with confidence. However,
aim of this study was to assess the stability of common centrifugation prior to transport is time consuming, and the
biochemical analytes in venous blood under our routine transfer of plasma or serum from a primary tube to a sec-
transport conditions and to evaluate analyte stability after ondary tube increases the risk of preanalytical errors. For
prompt or delayed centrifugation. analytes that are stable in whole blood for 24 h or more, it
Methods: We investigated pre- and postanalytical contri- seems that there is no benefit to centrifuge before transport.
butions of 32 biochemical analytes in plasma and serum
Keywords: pre- and postanalytical conditions; stability;
samples from 10 patients (healthy adults and patients
storage temperature; storage time; transport.
from intensive care units). Differences in analyte con-
centrations between baseline (T0) and different time
intervals (2, 4, 6, 8, 12 and 24  h) following storage after
prompt and delayed centrifugation were reported. Evalu-
Introduction
ation was against the total change limit as described by
A common problem with blood samples is maintaining
Oddoze et al. (Oddoze C, Lombard E, Portugal H. Stability
stability of analytes during tube transport from clinical
study of 81 analytes in human whole blood, in serum and
to pathology departments, and after centrifugation once
in plasma. Clin Biochem 2012;45:464–9).
they have arrived in the laboratory. Previous studies did
Results: The majority of analytes were stable with delayed
not meet all the criteria for use by all laboratories, either
separation up to 12  h, except for potassium, C-peptide,
because they were not complete or because they described
osteocalcin, parathyroid hormone (PTH), bicarbonate and
conditions of storage which did not reflect the usual prac-
tice of all laboratories. Preanalytical steps are especially
crucial when blood testing is carried out at a distance from
*Corresponding author: Jean Paul Cristol, Department of Biochemistry,
Lapeyronie University Hospital, 371 Av. Doyen G. Giraud,
clinical departments, leading to the transport of samples
34295 Montpellier, France, E-mail: [email protected]; and thus to associated variable environmental condi-
and PhyMedExp, INSERM U1046, CNRS UMR 9214, University of tions (such as temperature). More importantly, commonly
Montpellier, Montpellier, France required analytes (e.g. potassium) are known to be sen-
Anne Marie Dupuy, Mickael Mendes and Pascal Philibert: Department sitive to delayed centrifugation and temperature. In our
of Biochemistry, Lapeyronie University Hospital, Montpellier, France
daily practice, samples are transported in specimen boxes
Bruno Vincent: Intensive Care Medicine Department, Lapeyronie
University Hospital, Montpellier, France by the hospital courier service from clinical departments
Anne Sophie Bargnoux and Stéphanie Badiou: Department of to the biochemistry laboratory. Once in the laboratory, the
Biochemistry, Lapeyronie University Hospital, Montpellier, France; samples are centrifuged, analyzed in the primary tube and
and PhyMedExp, INSERM U1046, CNRS UMR 9214, University of then stored at 4 °C for 72 h. The aim of this study was to
Montpellier, Montpellier, France
determine the stability of common biochemical analytes
Kadda Klouche: PhyMedExp, INSERM U1046, CNRS UMR 9214,
University of Montpellier, Montpellier, France; and Intensive Care
in venous blood under our routine conditions. We evalu-
Medicine Department, Lapeyronie University Hospital, Montpellier, ated the analyte stability after prompt or delayed centrifu-
France gation at room temperature and with prompt or delayed

Brought to you by | University of Georgia Libraries

Authenticated
Download Date | 9/10/17 1:31 PM
2      Dupuy et al.: Stability of routine biochemical parameters

measurement with storage of primary tubes at 4 °C, to Methods

test if reanalysis is possible or if it is acceptable to allow
requests to be added on to the primary tube samples. The studied analytes corresponded to those for which data were
missing in the literature, either because the storage time described
in the publications was not long enough in relation to our practice
or to check stability of analytes known to be unstable or inconsist-
Materials and methods ent through a literature review. Also considered were those analytes
that are frequently added on by clinicians after obtaining initial
results – especially in emergency care units. The following analytes
Subjects and blood sampling
were studied from lithium heparin at each time point described
above: bicarbonate, ferritin, LDH, magnesium, NT-proBNP, phos-
The study was carried out in December 2015 in two series according phorus, potassium, total bilirubin and high-sensitivity troponin T.
to the type of tube: the first series was with lithium heparin vacuum The stability of the following analytes known to be stable for 24 h
tubes and the second with serum vacuum tubes. The study group [1] was tested only on tubes stored for 48  h after centrifugation:
consisted of six healthy volunteers (physicians) and four adult vol- alkaline phosphatase, alanine aminotransferase, aspartate ami-
unteers from intensive care units (Lapeyronie Hospital, Montpellier). notransferase, urea, calcium, chloride, creatine kinase, creatinine,
We chose healthy volunteers as well as hospitalized patients to cover C-reactive protein, gamma-glutamyl transferase, lipase, myoglobin,
both normal and pathological levels. The protocol was performed sodium and uric acid. From the serum tubes, nine analytes were
according to the principles of the Declaration of Helsinki, approved quantified: apolipoprotein A1, apolipoprotein B, cholesterol, HDL
by the Ethic Committee of Montpellier. Each donor was informed of cholesterol and triglycerides, as well as hormones C-peptide, insu-
the aim of the study and signed a consent form. For each person, lin, osteocalcin and parathyroid hormone (PTH). All assays were
seven samples were collected into 6-mL plastic Vacuette© lithium performed on a Cobas 8000© system including modules c701 and
heparin tubes, and seven samples were collected into plastic serum c502 (for biochemical analytes) and e602 (for immunochemistry)
tubes (Becton Dickinson [BD] Vacutainer©) with reference numbers from Roche Diagnostics (Meylan, France) using Roche reagent kits
368886 and 368815, respectively. After collection, the reference sam- and calibrators.
ple (T0 Ref, baseline) from each volunteer was centrifuged at 2000 g
for 10 min at 20 °C according to BD’s recommendations. The mean of
the results was calculated and represented the ‘T0 Ref’ or baseline for
each analyte. Serum was allowed to clot for 30 min at room tempera-
Statistical analysis
ture (between 20 °C and 25 °C) before centrifugation. The other tubes
were then stored at room temperature and gently inverted approxi- For the 10 healthy and pathological donors, results for each analyte
mately every 20 min to simulate transport by courier. They were then from the promptly centrifuged samples were considered as the ref-
centrifuged at different time intervals (2, 4, 6, 8, 12 and 24 h) and then erence and called ‘T0 Ref’ (baseline), expressed as the mean (min-
analyzed. After analysis, the primary tubes were stored at 4 °C (our max) in the Tables. For each storage time and for each analyte, the
routine daily conditions) and reanalyzed at different time intervals absolute difference of each sample (or percentage deviation) from the
(2, 4, 6, 8, 12, 24 and 48 h). F
­ igure 1 represents the design of the study T0 Reference concentration was calculated by the following formula:
organization. [(Tx − T0  Ref)/T0 Ref] × 100. Then the mean percentage deviation

Time period before

centrifugation at RT Reanalysis with storage at +4 °C
between each measure

T2 h T4 h T6 h T8 h T12 h T24 h T48 h

H1+S1 A
T0 Ref
A T2 h T4 h T6 h T8 h T12 h T24 h T48 h
H2+S2 T2 h
A T2 h T4 h T6 h T8 h T12 h T24 h T48 h
H3+S3 T4 h
For each donor (n = 10)
7 heparin (H) T6 h A T2 h T4 h T6 h T8 h T12 h T24 h T48 h
H4+S4
and 7 dry (S) tubes
were collected A T2 h T4 h T6 h T8 h T12 h T24 h T48 h
H5+S5 T8 h
A T2 h T4 h T6 h T8 h T12 h T24 h T48 h
H6+S6 T12 h

H7+S7 T24 h A T2 h T4 h T6 h T8 h T12 h T24 h T48 h

A: analysis; RT: room temperature; T0 Ref : T0 Reference (with a delay of 30 min for dry tubes)
: centrifugation of tubes.

Figure 1: Flow diagram of the study.

Brought to you by | University of Georgia Libraries

Authenticated
Download Date | 9/10/17 1:31 PM
 Dupuy et al.: Stability of routine biochemical parameters      3

(±SD) of the 10 donors was calculated. To assess the stability of the these analytes in centrifuged tubes cannot be allowed.
analytes, we calculated the total change limit (TCL) described by Phosphorus and potassium may be additionally requested
Oddoze et  al. [1] using the formula ­[(2.77 × CVa)2]½ + [(0.5 × CVb)2]½,
up to 12  h after centrifugation if the sample is stored at
where CVa is analytical imprecision and CVb is within-subject varia-
tion. If the mean percentage deviation (±SD) was higher than the TCL
4 °C and if the delay in transporting the blood is minimal.
value calculated for each analyte, the delay was considered unaccep- For hormonal analytes such as C-peptide, osteocalcin and
table and the analyte considered not stable. PTH, the mean concentrations exceeded TCL beyond 6 h.
However, the majority of analytes were stable in plasma or
serum for more than 48 h, even if the transport exceeded
2 h. Therefore, additional analysis or reanalysis could be
Results performed within 48 h of specimen collection.

Table 1 reports the stability of 18 biochemical and immu-

nological analytes when stored as whole blood (using
lithium heparin or serum tubes prior to centrifugation) Discussion
at room temperature. Table 2 shows the stability of these
same 18 analytes with measurements performed from 2 In this study, we investigated the influence of time and
to 48 h after prompt centrifugation on samples stored in temperature on the stability of analytes subject to delayed
primary tubes at 4 °C, as well as the stability at only 48 h centrifugation and after centrifugation under our daily
for 14 additional analytes. In Table 3, we report the stabil- routine conditions.
ity of analytes in samples stored in primary tubes at 4 °C Time before centrifugation leads to a major analytical
after the maximum acceptable delay of centrifugation. change in concentration for potassium, and minor analyti-
cal change for hormones, whereas storage time after cen-
trifugation strongly affects bicarbonate and LDH. Two main
Stability of analytes according to the time reasons explain the observed differences: first, the pro-
before centrifugation longed contact of plasma or serum with cells and release
of intracellular components (especially true for potas-
As shown in Table 1, using TCL as the goal, mean differ- sium, phosphorus, magnesium, LDH and insulin), and
ence exceeded the TCL after 4  h for plasma potassium, second, the storage temperature. At 20–25 °C, we observed
after 6  h for serum C-peptide, serum osteocalcin and a decrease then an increase in potassium level (J curve) and
serum PTH, after 8  h for plasma bicarbonate and LDH, an increase in potassium over a longer period at 4 °C (from
and after 12 h for plasma magnesium and phosphorus. All −1% to 9.6% between 2 and 24  h at 4 °C) due to leakage
other analytes were considered stable in primary tubes from cells. Proteins and peptides are susceptible to enzyme
stored for up to 24 h before centrifugation. degradation. Optimal stability for insulin, C-peptide and
PTH was observed when the samples were centrifuged and
stored at 4 °C (48, 48 and 24 h, respectively). The decrease
Stability in primary tubes stored at 4 °C in CO2 plasma concentrations (from −4% to 11% between
after prompt or maximum acceptable 2 and 4 h of storage) is probably due to exposure to air as
delayed centrifugation described by Kumar and Karon [2], namely, 1  mmol/L in
samples exposed to air for 1  h. Regarding LDH, in whole
Tables 2 and 3 summarize the results for plasma or serum blood the increase in concentration was due to exchange
for all analytes and the mean differences for different between plasma and blood cells. We also found an increase
storage times at 4 °C. Data for the analytes under refer- in plasma LDH concentration after centrifugation because
ence conditions (with centrifugation and immediate anal- cells remaining in plasma (even after centrifugation) lead
ysis after blood sampling, and a delay of 30  min before to release of LDH, similarly for potassium and phosphorus.
centrifugation for serum tubes), so-called T0 Reference Previous studies reported the influence of preana-
(T0 Ref, baseline), are reported in Table 2. In Table 3, we lytical conditions such as temperature, time and the type
report the data obtained for plasma/serum compared to of tube [1, 3–8]. However, few studies have reported the
the maximum acceptable delay before centrifugation of stability before and after centrifugation over such a long
each analyte (T6 h, T8 h, T12 h or T24 h). period of time. The results for hormones are consistent
LDH and bicarbonate were the analytes with the lowest with those of Oddoze et al. [1], who suggested collection
stability after centrifugation; therefore, any reanalysis of into K3EDTA to enhance stability in whole blood up to 72 h.

Brought to you by | University of Georgia Libraries

Authenticated
Download Date | 9/10/17 1:31 PM
 Table 1: Stability of biochemical analytes at room temperature (20–25 °C) according to delay before centrifugation.

Parameter   T0 Refa  Tube   TCL  Mean % difference at RT (±SD)   Acceptable

type   delay, h
T2 h   T4 h   T6 h   T8 h   T12 h   T24 h

Bicarbonate, mmol/L   25.8 (19–32)  LiH   ±8.6  −4.2 (4.0)   −5.3 (3.7)   −5.7 (3.3)   −6.9 (2.9)   −8.8 (3.7)   −14.0 (4.0)  8
Ferritin, ng/mL   689 (16–3301)  LiH   ±10.4  1.5 (2.7)   −0.1 (3.7)   −0.2 (1.6)   0.7 (1.9)   1.6 (2.2)   3.8 (3.5)  24
LDH, UI/L   159 (133–221)  LiH   ±8.8  −1.7 (2.0)   2.6 (5.9)   4.8 (6.2)   3.6 (6.6)   12.6 (19)   13.4 (15.7)  8
Magnesium, mmol/L   0.8 (0.7–0.9)  LiH   ±5.9  0.6 (1.8)   −0.4 (1.7)   1.0 (1.2)   1.2 (1.6)   2.7 (1.6)   7.7 (5.9)  12
NT-proBNP, pg/mL   444 (23–1761)  LiH   ±18.3  NT   −2.6 (2.0)   −2.2 (2.4)   −2.5 (3.0)   −3.4 (3.2)   −0.9 (2.7)  24
Phosphorus, mmol/L   0.9 (0.6–1.1)  LiH   ±6.8  −2.0 (2.6)   −4.5 (3.2)   −4.7 (5.5)   −5.3 (5.5)   0.3 (6.7)   58.4 (60)  12
4      Dupuy et al.: Stability of routine biochemical parameters

Potassium, mmol/L   3.8 (3.1–4.7)  LiH   ±4.8  −3.2 (1.6)   −4.5 (1.2)   −6.5 (2.7)   −7.9 (3.2)   −8.0 (4.7)   −3.6 (3.8)  4
Total bilirubin, μmol/L   8.2 (4–13)  LiH   ±20.5  −2.5 (11)   0.7 (8.2)   −0.9 (6.0)   −0.9 (6.0)   −1.2 (8.8)   −4.2 (9.0)  24
Troponin T, pg/mL   45.4 (<3–166.8)  LiH   ±9.3  NT   −1.7 (3.0)   −2.5 (3.8)   −2.2 (1.5)   −0.5 (5.5)   0.5 (4.9)  24
Total cholesterol, mmol/L   4.9 (3.4–6.8)  S   ±9.7  NT   NT   NT   NT   NT   4.3 (1.1)  24
TG, mmol/L   1.2 (0.7–2.2)  S   ±19.1  NT   NT   NT   NT   NT   6.3 (2.9)  24
HDL cholesterol, mmol/L   1.4 (0.9–1.7)  S   ±10.9  NT   NT   NT   NT   NT   3.2 (3.7)  24
Apolipoprotein A1, g/L   1.6 (1.3–1.8)  S   ±9.8  NT   NT   NT   NT   NT   −0.1 (3.2)  24
Apolipoprotein B, g/L   0.9 (0.7–1.5)  S   ±13.3  NT   NT   NT   NT   NT   −1.6 (1.5)  24
C-peptide, ng/mL   3.4 (1.8–6)  S   ±12.2  −2.2 (2.8)   −3.7 (3.6)   −6.1(2.1)   NT   NT   −23.4 (4.0)  6
Insulin, μU/mL   22.6 (6.3–57.8)  S   ±32.8  −1.8 (11.8)   −3.2 (13.1)   −7.5 (5.8)   NT   NT   −21.0 (9.8)  24
Osteocalcin, ng/mL   26.1 (16.2–35.4)  S   ±16.1  −4.4 (2.0)   −7.7 (2.3)   −12.1 (2.8)   NT   NT   −33.5 (6.2)  6
PTH, pg/mL   37.4 (20.9–57.6)  S   ±14.6  NT   NT   −10.6 (5.0)   NT   NT   −36.2 (13.8)  6
a
Reference sample expressed as mean (min–max); TCL, total change limit; RT, room temperature; LiH, lithium heparin; S, serum; NT, not tested; gray boxes: mean percentage higher than the TCL
value; TG, triglycerides; PTH, parathyroid hormone.

Download Date | 9/10/17 1:31 PM

Authenticated
Brought to you by | University of Georgia Libraries
 Table 2: Stability of biochemical analytes from reference samples in plasma (A) or serum (B) in centrifuged primary tubes stored at +4 °C.

Analyte   T0 Refa  Tube   TCL  Time, h   Mean % difference at +4 °C (±SD)   Acceptable
type   delay, h
T2 h   T4 h   T6 h   T8 h   T12 h   T24 h   T48 h

(A)                        
Bicarbonate, mmol/L   25.8 (19–32)  LiH   ±8.6  T0 Refa   −4.1 (1.3)   −11.3 (2.1)   −15.0 (2)   −19.7 (1.5)   −23.4 (3.3)   −27.6 (4.4)   −35.6 (5.9)   2
Ferritin, ng/mL   689 (16–3301)  LiH   ±10.4  T0 Refa   0.8 (2.5)   1.5 (1.7)   −1.1 (1.7)   −0.7 (1.1)   0.9 (2.1)   1.8 (2.5)   2.0 (4.6)   48
LDH, UI/L   159 (133–221)  LiH   ±8.8  T0 Refa   11.5 (5.4)   13.7 (7.1)   14.4 (9.1)   17.3 (10.6)   28.3 (11.0)   32.3 (10.8)   38.2 (13.1)   –
Magnesium, mmol/L   0.8 (0.7–0.9)  LiH   ±5.9  T0 Refa   0.1 (1.4)   1.0 (2.0)   1.9 (1.4)   2.2 (2.0)   3.1 (2.2)   5.2 (1.7)   5.0 (1.3)   48
NT-proBNP, pg/mL   444 (23–1761)  LiH   ±18.3  T0 Refa   −1.7 (3.3)   −1.0 (2.9)   −1.9 (2.5)   −1.0 (0.8)   1.1 (3.1)   2.0 (2.7)   1.0 (2.8)   48
Phosphorus, mmol/L   0.9 (0.6–1.1)  LiH   ±6.8  T0 Refa   1.9 (2.0)   1.9 (1.4)   2.8 (1.6)   3.3 (2.8)   4.1 (2.7)   7.7 (8.1)   16.2 (16.1)   12
Potassium, mmol/L   3.8 (3.1–4.7)  LiH   ±4.8  T0 Refa   −1.0 (1.3)   −0.3 (1.4)   0.8 (1.3)   1.2 (1.3)   2.1 (1.7)   9.6 (3.2)   36.5 (13.7)   12
Total bilirubin, μmol/L  8.2 (4–13)  LiH   ±20.5  T0 Refa   −0.6 (9.4)   −4.6 (10.0)   0.7 (8.2)   0.1 (6.9)   −2.0 (6.8)   −3.2 (7.5)   −2.1 (8.8)   48
Troponin T, pg/mL   45.4 (<3–166.8)  LiH   ±9.3  T0 Refa   −0.6 (0.8)   0.0 (0.6)   −1.0 (1.0)   −1.0 (1.2)   3.5 (5.6)   1.3 (1.8)   −0.8 (4.3)   48
ALP, UI/L   80.6 (33–117)  LiH   ±14.3  T0 Refa   NT   NT   NT   NT   NT   NT   0.4 (2.1)   48
ALT, UI/L   54 (14–275)  LiH   ±21.7  T0 Refa   NT   NT   NT   NT   NT   NT   −0.2 (2.5)   48
AST, UI/L   36 (11 to –129)  LiH   ±10.8  T0 Refa   NT   NT   NT   NT   NT   NT   14.7 (9.8)   48
BUN, mmol/L   7 (3–14.2)  LiH   ±12.0  T0 Refa   NT   NT   NT   NT   NT   NT   5.7 (1.5)   48
Calcium, mmol/L   2.22 (1.90–2.40)  LiH   ±4.3  T0 Refa   NT   NT   NT   NT   NT   NT   4.4 (1.0)   48
Chloride, mmol/L   103.8 (97–110)  LiH   ±5.5  T0 Refa   NT   NT   NT   NT   NT   NT   −0.8 (0.8)   48
CK, UI/L   518 (57–2362)  LiH   ±21.0  T0 Refa   NT   NT   NT   NT   NT   NT   −0.0 (4.5)   48
a
Creatinine, μmol/L   86.6 (34–262)  LiH   ±10.1  T0 Ref   NT   NT   NT   NT   NT   NT   4.4 (1.6)   48
CRP, mg/L   87.8 (<0.3–340)  LiH   ±38.7  T0 Refa   NT   NT   NT   NT   NT   NT   4.2 (5.1)   48
GGT, UI/L   78.9 (10–376)  LiH   ±20.9  T0 Refa   NT   NT   NT   NT   NT   NT   5.8 (3.1)   48
Lipase, UI/L   31.4 (6–48)  LiH   ±17.4  T0 Refa   NT   NT   NT   NT   NT   NT   2.2 (2.1)   48
Myoglobin, μg/L   274.7 (27.9–1146)  LiH   ±16.0  T0 Refa   NT   NT   NT   NT   NT   NT   −3.6 (4.6)   48
Sodium, mmol/L   141 (136–144)  LiH   ±3.6  T0 Refa   NT   NT   NT   NT   NT   NT   2.1 (1.8)   48
Uric acid, μmol/L   238 (221–389)  LiH   ±9.6  T0 Refa   NT   NT   NT   NT   NT   NT   3.5 (1.2)   48

Download Date | 9/10/17 1:31 PM

Authenticated
Brought to you by | University of Georgia Libraries
Dupuy et al.: Stability of routine biochemical parameters      5
 6      Dupuy et al.: Stability of routine biochemical parameters

For the majority of classical biochemical analytes, the sta-

Acceptable
delay, h

48
48
48
48
48
48
48
6
24
bility reported in our study was in accordance with other
studies and even longer (up to 48  h). However, for AST,

Reference sample expressed as mean (min–max). TCL, total change limit; RT, room temperature; LiH, lithium heparin; S, serum; NT, not tested; ALP, alkaline phosphatase; ALT, alanine
in our study the stability in plasma was <48  h. In prac-

aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CK, creatine kinase; CRP, C-reactive protein; GGT, gamma-glutamyl transferase; TG, triglycerides; PTH,
  tice, LDH, magnesium, phosphorus and potassium are
 

3.5 (1.2)  
7.6 (4.7)  
−1.0 (2.6)  
2.1 (3.1)  
−2.1 (2.3)  
−10.1 (2.5)  
−11.2 (2.5)  
 
 
the analytes most influenced by delayed centrifugation,
T48 h

as they are present in cells. The stability of magnesium,

NT
NT
phosphorus and LDH was the longest in our study at 12, 12
and 8 h, respectively, compared to other studies [1, 3–8],
 

 
2.4 (1.2)  
4.5 (2.7)  
−2.9 (1.3)  
 
−1.2 (2.2)  
−3.7 (1.4)  
−5.3 (2.3)  
 
−11.0 (3.1)  
whereas the result for bicarbonate was in agreement with
0.7 4.6)
T24 h

that of Oddoze et al. [1]. The tendency in the field of bio-

NT
chemistry is the centralization of analysis on automated
platforms, so it is important to evaluate if the samples
 

 
 
 
 
 
 
 
 
 
−10.8 (4.0)  

need to be centrifuged before or after transport to the plat-

T12 h

form. This question arises, in particular, for the most sen-

NT
NT
NT
NT
NT
NT
NT
NT

sitive analytes required by clinicians who are poorly aware

of the effect conditions have on stability. We found, and
T8 h  

 
 
 
 
 
 
 
 
 
 
NT
NT
NT
NT
NT
NT
NT
NT
NT

this was verified in previous studies, that once the tube is

centrifuged (stored and then analyzed from the primary
 

 
 
 
 
 
 
−1.2 (0.8)  
−2.3 (1.0)  
−8.1 (1.8)  
−10.3 (3.4)  

tube or from an aliquot of plasma), the stability of the ana-

lytes is less affected. For example, plasma potassium was
T6 h
Mean % difference at +4 °C (±SD)

NT
NT
NT
NT
NT

found to be stable up to 12 h when the sample was centri-

fuged and stored at 4 °C. Potassium, as expected, was the
 

 
 
 
 
 
 
−1.6 (0.6)  
−2.1 (1.3)  
−5.1 (1.5)  
 

least stable analyte in whole blood (prior to centrifuga-

tion) and did not remain stable beyond 4 h after sampling.
T4 h

NT
NT
NT
NT
NT

NT

In addition, according to previous studies, temperature is

an essential factor in stability. In our hospital, the effect
 

 
 
 
 
 
 
−0.6 (0.9)  
−2.0 (1.8)  
−4.4 (1.7)  
−10.6 (3.8)  

of temperature change was negligible as samples were

not subject to external variations, unlike some platforms
T2 h

NT
NT
NT
NT
NT

that are located at a distance from sampling sites. Several

studies reported different data on the stability of potas-
parathyroid hormone; gray boxes: mean percentage higher than the TCL value.
 
Time, h  
 

 
 
 
 
 
 
 
 
 
T0 Refa
T0 Refa
T0 Refa
T0 Refa
T0 Refa
T0 Refa
T0 Refa
T0 Refa
T0 Refa

sium in whole blood as a function of transport, storage

conditions and type of tube. In Table 4, we have summa-
rized the main publications that studied various stability
 
±9.7 
TCL 

±19.1 
±10.9 
±9.8 
±13.3 
±12.2 
±32.8 
±16.1 
±14.6 

conditions for potassium, particularly using heparinized

tubes (with or without gel). With a mean temperature of
 
 
Tube  

 
 
 
 
 
 
 
 

23 °C, our results were in agreement with those of Stahl

type

and Brandslund [8]. In conclusion, it is clear that deter-

S
S
S
S
S
S
S
S
S

mining the stability of analytes is crucial to interpret bio-

 
4.9 (3.4–6.8) 
T0 Refa 

1.2 (0.7–2.2) 
1.4 (0.9–1.7) 
1.6 (1.3–1.8 
0.9 (0.7–1.5) 
3.4 (1.8–6) 
22.6 (6.3–57.8) 
26.1 (16.2–35.4) 
37.4 (20.9–57.6) 

chemical testing with confidence. Centrifugation before

transport is time consuming, and the transfer of plasma or
serum from a primary tube to a secondary tube increases
the risk of preanalytical errors. For those analytes that are
stable in whole blood for 24 h or more, it seems that it is not
necessary to centrifuge before transport. To improve the
 
Total cholesterol, mmol/L 
 

 
HDL cholesterol, mmol/L  
Apolipoprotein A1, g/L  
 
 
 
 
 

stability of analytes, tube manufacturers such as Becton

Apolipoprotein B, g/L
Table 2 (continued)

Osteocalcin, ng/mL

Dickinson or Greiner have developed several options. One

C-peptide, ng/mL

option is to use dry or lithium-gel separation tubes that

Insulin, μU/L

PTH, pg/mL
TG, mmol/L

allow better stability of some analytes if the centrifuga-

Analyte

tion is carried out quickly after sampling. Another option

is the existence of serum tubes with a combination of two
(B)

Brought to you by | University of Georgia Libraries

Authenticated
Download Date | 9/10/17 1:31 PM
 Table 3: Stability of biochemical analytes in plasma (A) or serum (B) in centrifuged primary tubes stored at +4 °C with the maximum acceptable delay before centrifugation for each parameter.

Analyte   T0 Refa  Tube   TCL  Time,   T0  Mean % difference at +4 °C (±SD)   Acceptable
type h   delay, h
T2 h   T4 h   T6 h   T8 h   T12 h   T24 h   T48 h

(A)                          
Bicarbonate, mmol/L   25.8 (19–32)  LiH   ±8.6  T8 h   −6.9 (2.9)  −5.3 (2.3)  −11.9 (2.3)  −16.3 (3.1)  −18.8 (3.1)  −22.6 (3.7)  −27.1 (4.1)   −36.4 (5.8)  2b
Ferritin, ng/mL   689 (16–3301)  LiH   ±10.4  T24 h   3.8 (3.5)  −2.1 (2.1)  −0.2 (3.3)   1.1 (1.6)   1.9 (2)   2.2 (2.4)   4.0 (3.3)   NT   24b
LDH, UI/L   159 (133–221)  LiH   ±8.8  T8 h   3.6 (6.6)  10.6 (8.1)   11.6 (5.6)   13.7 (5.6)   16.5 (6.7)   25.6 (8.7)   24.3 (6.6)   37.6 (15.8)  –
Magnesium, mmol/L   0.8 (0.7–0.9)  LiH   ±5.9  T12 h   2.7 (1.6)  0.5 (1.4)   2.2 (1.2)   3.2 (1.5)   2.9 (1.3)   3.4 (2.4)   3.9 (2.1)   6.38 (5.3)   24b
NT-proBNP, pg/mL   444 (23–1761)  LiH   ±18.3  T24 h   −0.9 (2.7)  −1.0 (1.9)  −1.7 (1.4)   −0.7 (1.3)   −0.1 (1.3)   0.0 (2.9)   1.7 (3.7)   NT   24b
Phosphorus, mmol/L   0.9 (0.6–1.1)  LiH   ±6.8  T12 h   0.3 (6.7)  0.9 (1.7)   2.3 (1.1)   3.3 (2.1)   6.0 (1.5)   9.0 (10)   10.7 (7.2)   26 (23)   8b
Potassium, mmol/L   3.8 (3.1–4.7)  LiH   ±4.8  T4 h   −4.5 (1.2)  0.8 (1.3)   0.5 (1.1)   1.3 (1.4)   2.2 (1.7)   3.0 (0.9)   10.3 (3.5)   37.6 (11)   –
Total bilirubin, μmol/L   8.2 (4–13)  LiH   ±20.5  T24 h   −4.2 (9)  2.6 (8.1)   −0.2 (4.8)   5.5 (10)   2.3 (9.3)   −1.1 (7.7)   1.1 (10)   NT   24b
Troponin T, pg/mL   45.4 (<3–166.8)  LiH   ±9.3  T24 h   0.5 (4.9)  0.9 (3.2)   0.3 (2.2)   0.7 (1.1)   0.9 (1.3)   1.7 (5.7)   −1.2 (6.2)   NT   24b
                         
(B)                          
Total cholesterol, mmol/L   4.9 (3.4–6.8)  S   ±9.7  T24 h   4.3 (1.1)  NT   NT   NT   NT   5.9 (1.5)   8.0 (1.7)   NT   24b
TG, mmol/L   1.2 (0.7–2.2)  S   ±19.1  T24 h   6.3 (2.9)  NT   NT   NT   NT   7.6 (3.5)   9.7 (4.3)   NT   24b
HDL cholesterol, mmol/L   1.4 (0.9–1.7)  S   ±10.9  T24 h   3.2 (3.7)  NT   NT   NT   NT   5.1 (4.5)   7.8 (2.4)   NT   24b
Apolipoprotein A1, g/L   1.6 (1.3–1.8)  S   ±9.8  T24 h   −0.1 (3.1)  NT   NT   NT   NT   4.4 (4.6)   7.1 (3.1)   NT   24b
Apolipoprotein B, g/L   0.9 (0.7–1.5)  S   ±13.3  T24 h   −1.6 (1.5)  NT   NT   NT   NT   −0.9 (2.6)   0.9 (2.8)   NT   24b
C-peptide, ng/mL   3.4 (1.8–6)  S   ±12.2  T6 h   −6.1 (2.1)  NT   NT   NT   NT   NT   NT   −15.1 (2.3)  –
Insulin, μU/L   22.6 (6.3–57.8)  S   ±32.8  T24 h   −21 (9.8)  NT   NT   NT   NT   NT   NT   −12.6 (5.9)  48b
Osteocalcin, ng/mL   26.1 (16.2–35.4)    ±16.1  T6 h   −12.1 (2.8)  NT   NT   NT   NT   NT   NT   −22.2 (2.4)  –
PTH, pg/mL   37.4 (20.9–57.6)  S   ±14.6  T6 h   −10.6 (8)  NT   NT   NT   NT   −24.9 (11)   −24.0 (11.6)  −29.4 (12)   –
a
Reference sample expressed as mean (min–max); bacceptable delay for reanalysis taking into account the maximum acceptable delay before centrifugation for each parameter; TCL, total
change limit; RT, room temperature; LiH, lithium heparin; S, serum; NT, not tested; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CK, creatine
kinase; CRP, C-reactive protein; GGT, gamma-glutamyl transferase; TG, triglycerides; PTH, parathyroid hormone; gray boxes, mean percentage higher than the TCL value.

Download Date | 9/10/17 1:31 PM

Authenticated
Brought to you by | University of Georgia Libraries
Dupuy et al.: Stability of routine biochemical parameters      7
 Table 4: Main publications on the stability of potassium from lithium heparin tubes.

Reference   Type of  n  Preanalytical conditions   Goal used   Acceptable delay

tube    
Storage time, h  Storage temperature, °C Stability before   Stability in plasma
centrifugation

Boyanton and Blick [3]   PST   10  0, 4, 8, 16, 24, 32, 40, 48, 56  RT (25)   Significant change limit   <4 h at 25 °C   NT
Stahl and Brandslund [8]   LiHep   5  0, 2, 4, 6, 8  17, 20, 23 and 25   Analytical bias [9, 10]   8–12 h at 20 °C   NT
            4 h at 23 °C  
            <2 h at 25 °C  
            <2 h at 17 °C  
O’Keane and Cunningham [11]  LiHep   56  0, 24, 48  RT (20)   Total allowable error   <24 h at 20 °C   T48 h if separation
8      Dupuy et al.: Stability of routine biochemical parameters

of plasma/cells
and stored at 4 °C
Jensen et al. [5]   LiHep   406  0, 4, 6, 8  Transport of samples in winter with   Allowed deviation   6 h at 20–25 °C   NT
PST outdoor temperatures of −8.6, 1.7 and 10 defined by authors
        Transport of samples in summer with     8 h at 20–25 °C  
outdoor temperatures of 21, 26 and >30
Leino and Koivula [6]   PST   50  6  8, 22   Significant change limit   <6 h at 8 °C   NT
  6 h at 22 °C  
Oddoze et al. [1]   LiHep   50  0, 2, 4, 6, 24  4, 25   Total change limit   <2 h at 4 °C   >4 h at 4 °C
  <2 h at 25 °C   >4 h at 25 °C
Henriksen et al. [4]   LiHep   106  0, 10  21 ± 1   Goal bias and analytical-   10 h at 21 °C   NT
and goal-CV [12]
Monneret et al. [7]   PST   28  3, 4, 5, 6  RT (21.3 ± 1.8)   Acceptable change limit   +5 to +6 h   >6 h
Present study   LiHep   10  0, 2, 4, 6, 8, 12, 24  RT (20–25)   Total change limit   4 h at 20–25 °C   T12 h at 4 °C

RT, room temperature; LiHep, lithium heparin tubes without gel; PST, plasma separator gel tubes containing lithium heparin; n, number of donors included in the study.

Download Date | 9/10/17 1:31 PM

Authenticated
Brought to you by | University of Georgia Libraries
 Dupuy et al.: Stability of routine biochemical parameters      9

procoagulants, blood clotting activator and thrombin, 3. Boyanton BL Jr, Blick KE. Stability studies of twenty-four ana-
resulting in a significant decrease of the clotting time from lytes in human plasma and serum. Clin Chem 2002;48:2242–7.
4. Henriksen LO, Faber NR, Moller MF, Nexo E, Hansen AB. Stabil-
30 min to <5 min. In consequence, centrifugation may be
ity of 35 biochemical and immunological routine tests after
carried out quickly after the blood sample is taken, allow- 10 hours storage and transport of human whole blood at 21°C.
ing minimized cell contact and thus improving the stabil- Scand J Clin Lab Invest 2014;74:603–10.
ity of analytes [13]. Further studies must be carried out to 5. Jensen EA, Stahl M, Brandslund I, Grinsted P. Stability of heparin
confirm this result. The strength of the study is the inclu- blood samples during transport based on defined pre-analytical
quality goals. Clin Chem Lab Med 2008;46:225–34.
sion of pathological subjects, allowing better assessment
6. Leino A, Koivula MK. Stability of chemical and immunochemical
of the stability of values covering the whole measurement
analytes in uncentrifuged plasma samples. Ann Clin Biochem
range. 2009;46:159–61.
7. Monneret D, Godmer A, Le Guen R, Bravetti C, Emeraud C,
Acknowledgments: We gratefully acknowledge the bio- ­Marteau A, et al. Stability of routine biochemical analytes in
chemistry laboratory personnel for technical assistance. whole blood and plasma from lithium heparin gel tubes during
6-hr storage. J Clin Lab Anal 2016;30:602–9.
Author contributions: All the authors have accepted
8. Stahl M, Brandslund I. Controlled storage conditions prolong
responsibility for the entire content of this submitted stability of biochemical components in whole blood. Clin Chem
manuscript and approved submission. Lab Med 2005;43:210–5.
Research funding: None declared. 9. Gowans EM, Hyltoft Petersen P, Blaabjerg O, Hørder M. Analyti-
Employment or leadership: None declared. cal goals for the acceptance of common reference intervals for
laboratories throughout a geographical area. Scand J Clin Lab
Honorarium: None declared.
Invest 1988;48:757–64.
Competing interests: The funding organization(s) played
10. Fraser CG, Hyltoft Petersen P, Libeer JC, Ricos C. Proposals for
no role in the study design; in the collection, analysis, and setting generally applicable quality goals solely based on biol-
interpretation of data; in the writing of the report; or in the ogy. Ann Clin Biochem 1997;34:8–12.
decision to submit the report for publication. 11. O’Keane MP, Cunningham SK. Evaluation of three different
specimen types (serum, plasma lithium heparin and serum gel
separator) for analysis of certain analytes: clinical significance
of differences in results and efficiency in use. Clin Chem Lab
References Med 2006;44:662–8.
12. Rodbard D. Statistical quality control and routine data process-
1. Oddoze C, Lombard E, Portugal H. Stability study of 81 analytes ing for radioimmunoassays and immunoradiometric assays. Clin
in human whole blood, in serum and in plasma. Clin Biochem Chem 1974;20:1255–70.
2012;45:464–9. 13. Dimeski G, Johnston J, Masci PP, Zhao KN, Brown N. Evaluation
2. Kumar V, Karon BS. Comparison of measured and calculated of the Greiner bio-one serum separator BCA fast clot tube. Clin
bicarbonate values. Clin Chem 2008;54:1586–7. Chem Lab Med 2017;55:1135–41.

Brought to you by | University of Georgia Libraries

Authenticated
Download Date | 9/10/17 1:31 PM
View publication stats