Evaluation of a new quality control material for erythrocyte ...

lovinggudgeonMechanics

Feb 22, 2014 (3 years and 6 months ago)

56 views

CONTR
IBUTI
SC
I
ENT
IFICI






SCIENTIFIC PAPERS


BIOCHIMICA CLINICA

EXTRACT OF VOL. 23
-

NR. 6 NOVEMBER
-
DECEMBER 1999



-

1
-


EVALUATION OF A NEW QUALITY CONTROL MATERIAL FOR
ERYTHROCYTE SEDIMENTATION RATE


Paola Luraschi
l
, Anna Maria Morelli
l
, Simona Brambilla
2
, Carlo Franzini
2

l
Laboratorio di Analisi Chimico
-
Cliniche, Ospedale L. Sacco, Milano

2
Cattedra di Biochimica Clinic
a, Istituto di Scienze Biomediche Ospedale L. Sacco, Università degli Studi di
Milano, Milano


ABSTRACT


Evaluation of a new quality control material for erythrocyte sedimentation rate

In spite of its empirical nature, the measure of the erythrocyte sedi
mentation rate (ESR) is very
frequently used in clinical med
i
cine. The ESR is the "acute phase
slow
reactant " more often used in
med
i
cine. "Fast" markers (i.e. C
-
reactive protein) are not considered
not
to be alternative but
complementary to the "slow" ma
rkers: the coupled determination would allow to optimize the
achievable information. Among the
slow
markers alternative to the ESR, the measure of plasma
viscosity is proposed, even if poorly used in practice
;

in
analytical
terms it has the advantage to po
ssess
quality control materials, while ESR does not. The new material proposed for the quality control of
ESR exhibits characteristics suited to the expected scope. The availability of such a quality control
material contributes to adjust the analytical re
liability of ESR to its widespread clinical use.


INTRODUCTION


ESR

test
ing
,
introduced
almost 80 years ago,
is

still
frequently request
ed

(1):
its

high sensi
bility
,
as a

generic marker o
f "inflammation", compensates
its

aspecificit
y

against the "type of

disease" (2) and
against
its

empirical nature.
The
phenomenon of
erythrocyte

sedimentation
, and
its
ruling
mechanisms
, are not fully known (1);
contribution
to determine its entity
(3,4)

is given by
erythrocyte
and plasma factors
,
among which
the
change i
n
concentration
of
some plasma proteins (primarily
the fibrinogen)
which
meet the requirements for
their classification as "acute phase proteins


(4). For
this reason, and
due to
vast clinical evidence

accumulated
over

the past
eighty years
,
ESR

is

widely
used as marker of inflammation or as a
n

“acute phase reagent”

(4).

The opportunity to
continue using

this test

is
often
debated
,
fundamentally
on the basis of the

empirical
nature of the measure
ment
, although the analysis of
its

merits
compared
with
the
d
etermination of other
acute phase
reagents (proteins) usually does not lead
to this conclusion (5
-
7).
In
recent years

observations
have been reported

on
the clinical utility of this
test
in different situations (8, 9).

Much attention

is being
paid to the
possibility

of
replacing
ESR
determination with measure
ment of
the C
-
reactive (CRP) protein
,

less empirical and
better

standardizable

(
10
).

The
measure
ment
of this
protein
appears of undoubtedly great clinical
relevance
, especially in
prevailing acute cl
in
ical
situations

(11
-
15).
On the other hand
,
there are
several
remarks

on the non
-
identity
of
the
information provided by the two determinations
(
ESR
and CRP), which would
result
complementary
rather than a
lternative (16
-
22), also in the
elderly
people

path
ology

(23, 24). In principle, however,
it
i
s not possible
to
establish absolute
and
generally
valid

criteria

to
determine

the clinica
l superiority of
one measure
ment

(ESR
)
over the other
(CRP) (25,
26). The

additional measure
ment

of
acute phase
serum
prote
ins

may
instead
give

the opportunity to
modulate the
differential
diagnos
tic significance

of

high ESR values,

even
with
the application of
mathematic
al

models (27). In certain conditions
(viral infections,
transplant rejections
) another acute
phase protein
,

the
serum

amyloid
A
,

CONTR
IBUTI
SC
I
ENT
IFICI






SCIENTIFIC PAPERS


BIOCHIMICA CLINICA

EXTRACT OF VOL. 23
-

NR. 6 NOVEMBER
-
DECEMBER 1999



2

yet little used in practice
, would result

a
more
sensitive

marker than
CRP (28).
By

contrast, the
concentration of serum procalcitonin

would

represent a more
specific marker of
sepsis
/
infection

(29).

In

consideration

of the frequent

clinical use of
ESR
and
of
its

empirical
characteristics,
several
times
the international organizations involved in the
standardization of
clinical laboratory measurement

methods have
worked

on the standardization of
ESR as
the
unique
approach to
the
norm
al
ization of
results (1, 30
-
32), i
n
absence
of
calibration and
control materials
.

Under the analytical profile,
the lack of
control
material is in fact one of the weaknesses of
ESR
(
1
), for which
rather complex alternative control
methods
are

proposed (1)
. The availability of a
control
material is in fact a
point in favor

of

plasma

viscosity measurement (4), indicated as a possible,
valid alternative to ESR (3, 4,

22, 33), but little
used in practice.

The availability of
ESR
automated instruments has
cert
ainly contributed to its analytical reliability and
feasibility, especially in centralized laboratories.
However, in principle the more automated is a
method, the more urgent is the necessity to
perform
checks and quality controls. Recently some ESR
contro
l materials have been released for use on
these systems, thus able to complete the analytical
approach to the measurement. In this paperwork
are reported the

results of an evaluation of these
materials; preliminary data were presented at the
17
th

Congress
o lnternazionale di Chimica Clinica,
Florence, 6
-
11 June 1999 (34).


MATERIA
LS AND METHODS


28 routine samples of fresh whole blood, of which
12 with ESR values less than 15 mm and 16 with
values between 27 mm and 115 mm were used for
stability testing of
human material. The blood
samples were collected directly into ESR tubes
with the instrument in use; the ESR was performed
in duplicate 2
-
4 hours after withdrawal and
repeated 6 times during 8 days, on the same
samples stored at 4
°
C. The control material,
at
2

ESR levels (material I and II), was provided by
Diesse

(Siena, Italy), stored at 4°C and handled
according to manufacturer’s instructions. Every
day
,
two ESR tubes were prepared with each of the
two control materials accurately mixed and
balanced at r
oom temperature, performing ESR
immediately.

The procedure was repeated 35 times
during 66 days. All measurements were made in
duplicate using Diesse (Siena, Italia) automated
instrument VES
-
MATIC PC. The results were
processed using
common
statistical par
ametric
tests
.


RESULTS


The results
of whole blood samples stability
check
(12 with normal values and 16 with ESR increased
values) are shown in Figure 1. As clearly indicated,
the ESR values fall
down
sharply and randomly,
approaching to “zero” in abou
t one week, regardless
of their initial value.

Figure 2 shows the ESR values flow
as a
function of

storage
tim
e, measured in both control samples. For
both samples no significant value changes are
shown for the whole duration of the test (66 days).

Table

I

shows the statistical parameters of ESR
value regression measured in control samples
against storage time.

T
he reported data highlight the good stability of
tested control materials: the values of slope
correspond to a daily average increase of 0.06 mm

for material I (average 8.0


1.9 mm) and a
variation not significantly different from zero (
-
0.02
mm/day) for material II (average 69.0


2.9 mm).
The
imprecision

values within the ESR range in
control materials testing (at 2 ESR values) and in
fresh blo
od samples (at 3 ESR values)
c
alculated

from
duplicate analysis results
, are summarized in
T
able II. In figure III a
chart

of
imprecision

values
is reported at different levels against the average
value (
imprecision

profile
)
: note how the points
related to

control materials
find themselves

on the
same curve of the fresh blood samples.

CONTR
IBUTI
SC
I
ENT
IFICI






SCIENTIFIC PAPERS


BIOCHIMICA CLINICA

EXTRACT OF VOL. 23
-

NR. 6 NOVEMBER
-
DECEMBER 1999



3



Figure 1

ESR stability
check
in 28 whole blood fresh samples (12 with normal
values, lower
chart
; 16 with increased values, upper
chart
) stored at
4°C during 8 days.





Figure 2

ESR value stability

check
with the two
evaluated
control materials,
stored at 4°C during 66 days. The continuous and
the
dashed lines
indicate the average values and the intervals


2 standard deviations.


DISCUSSION


The acute phase reagent
s are blood substances or
properties

measurable under even “normal”
conditions,
which

increase
in response to an
inflammatory stimulus
. A systematic approach to
their classification consists of three groups (4) based
on different criteria including the
res
ponse speed

and normalization
after

appearance/disappearance
of the stimulus. The majority of Authors agree on
the opportunity to associate the measurement of a
“rapid” marker with a "slow" marker (4, 22, 35, 36)
in agreement with the several remarks prev
iously

discussed, concerning the complementarity of the
information provided, for example,
by
ESR and
CRP.

There is a substantial agreement on preferential use
of CRP as a rapid marker,

notwithstanding

some
reservations regarding viral infections. In fact
, not
only
CRP
increases and normalizations are
extremely rapid, but the values of concentration that
it reaches are fairly well correlated with the entity of
the inflammatory process and can also be extremely
high (4). All these factors contribute to infl
uence the
clinical relevance of the test, even for the purpose of
documenting the inflammatory

component in
pathologies of
etiopathogenesis

still under debate
(37).

As to what concerns the choice of the slow marker,
there is less agreement; although the u
se of some
specific proteins is suggested, for example

-
1
-
acid
glycoprotein, or better,

-
1
-
antitrypsin (35), since
many years, however, it is preferred to use ESR (4,
17, 18, 38). On this test there is a vast clinical
experience; two further obser
vations

play to its
advantage

as a "slow" marker for inflammation: the
kinetics of increase and normalization is even
slower than other specific proteins
;

in the case of
strong clinical severity ESR values may reach 7
-
8
times the superior reference limit (15 mm),

which
usually does not happen with other mentioned
specific proteins. The immunochemistry
measurement of serum

-
1
-
acid glycoprotein could
also be influenced by factors independent by its
effective concentration (39, 40).

A valid alternative to ESR would

be represented by
the measurement of plasma viscosity, the values of




Stor
age time at 4°C (days)

Storage time at 4°C (days)

E

S

R


m

m

Storage time at 4°C (days)

E

S

R


m

m

E

S

R


m

m

CONTR
IBUTI
SC
I
ENT
IFICI






SCIENTIFIC PAPERS


BIOCHIMICA CLINICA

EXTRACT OF VOL. 23
-

NR. 6 NOVEMBER
-
DECEMBER 1999



4

Table

I

Statistical parameters of (average) ESR values regression during the storage time (at 4°C) during days


Material

Number of tests (*)

Average


DS (mm)

Intercept (mm)

Slope

E
S

Material I

35

8,0

1,9

6,4

0,06

0,01

Material II

35

69,0

2,9

70,0

0,02

0,02 (#)

(*) During 66 days

(#) Not significantly different from zero


Table
II

ESR
Imprecision

in the control materials and in fresh blood samples


Sample

Average

(
mm
)

Imprecision

(CV%)

In
-
run

Betwee
n
-
run

(*)

Material I

8,0

12,8

23,7

Material II

69,0

2,0

4,3

Fresh blood (low)

9,6

14,4

(#)

Fresh blood
(average)

30,5

6,2

(#)

Fresh blood
(
high
)

82,2

2,8

(#)

(*) During 66 days

(#) Not
measurable due to sample instability





which are fairly well correlated to ESR ones (3, 4,
22, 33, 36), although
it seems there is not
a relation
of cause/effect between the two dimensions (3).
Although only a limited analytical and clinical
experience is available on this measurement, it
has
the advantage to be more easily calibrated and
controlled with suitable materials (4).

The data reported in this paperwork confirm the
already known instability of ESR values measurable
in whole blood samples stored at low temperature
(1), which makes

them unsuitable to be used as
calibrators and/or controls. By contrast, the
evaluated materials show a perfect stability for a
period of over two months’ time. To be
used as a

control, a material should possess additional
features, among which commutabili
ty (41). It is not
always easy to determine in details this last
property: the
imprecision

profile
here

reported
shows the
characteristic

to generate an
imprecision

of measurement comparable to the one of fresh
blood samples. When the commutability is not
determined or unknown, the generally used
alternative is to assign

to the material

system
-
specific

values,
that is
characteristic of any
analytical system. If the assignment of these last
values is done correctly, it is possible to
ascertain

that when the
system gets the
assigned
system
-
specific

value for control materials, the

s
ystem
itself
generates
for the assayed samples
values
superimposable to those of
the

reference method
chosen

(42)
,

thus ensuring the chain of traceability.

The reported data
deter
mine

the suitability of the
studied material to act as a
n

ESR control material.
The availability of such materials contributes to the
analytical reliability of this measure
ment

widely
used in clinical medicine.


THANKS

T
hanks to DIESSE company (Siena) f
or providing the materials used in this test.
For bibliography, please revise the original

ESR (mm)



Controls


Fresh blood

C

V


b

e

t

w

e

e

n


r

u

n

(
%)