![]()
34
| 2 | 2006
Klinisk Biokemi i Norden
35
| 2 | 2006
Klinisk Biokemi i Norden
hospital laboratory and in one of the primary care
centres. For the other primary care centre 10% of
measurements deviated even more than the goals
stated by Chempaq.
For
monocytes
the goal was fulfilled for the sam-
ples from hospital. Only between 62 and 80 % of
the results in primary care were within ±43%. The
main reason is probably a slight positive bias for
Chempaq results compared to results from Sysmex
SE 9000 and Coulter (~0.2x10
9
/L).
Eosinophilocytes
are counted as granulocytes by
the Chempaq, which therefore not can be used to
detect patients with a high number of eosinophi-
locytes.
Variant Lymphocytes:
(atypical lymphocytes, viro-
cytes) cannot be distinguished by Chempaq. The
system can therefore not be expected to detect
patients with mononucleosis infectiosa.
Flags:
Chempaq had less than 1% of flags in outpatients.
In the severely ill patients the laboratory had more
remarks in the differentials than Chempaq. In 36 of
142 selected samples the Sysmex system indicated in
duplicates the need for a manual differential counting.
In 18 of them Chempaq also had flags in duplicates, in
2 Chempaq had flags in one of the duplicates while 16
had no remarks in Chempaq. Of the latter 16 samples
9 were normal or had band neutrocytes, 2 had variant
lymphocytes. Among the selected ‘in hospital’ patients
7 samples had early stages of granulocytes in low
concentrations identified by flags in Sysmex, but were
without flags in Chempaq.
Errors (Waste percentage of cassettes)
An error rate of 1,8% in the first evaluation increa-
sed to 14% in the second evaluation due to a later
identified technical error.
The user friendliness
of the ‘Manual’, ‘Time fac-
tors’ and ‘Operation’ were rated ‘satisfactory’ by
the technologists in the hospital laboratory. The
primary care centres pointed out that the applica-
tion of the samples, especially the capillary ones
could be difficult. Apart from this remark both
centres found the user friendliness either ‘satis-
factory’ or ‘very satisfactory’. In primary care the
precision seemed to improve by training, since
the within sample precision for both haemoglobin
and Leukocytes improved for the last 20 samples
compared to the first 20 samples. This was the case
for both venous and capillary samples. A special
observation made during the study was that the
glue used to organize the paper printouts from the
printer (Tesa lim-stift, Beirsdorf, Spain) made the
prints vanish after 14 days. The complete Chempaq
evaluation report is available at www.SKUP.nu or
www.SKUP.dk
References
1. Kvalitetskrav og kvalitetsvurdering for
hyppigt udførte klinisk biokemiske og
klinisk mikrobiologiske analyser i almen
praksis. Konsensus dokument udarbejdet af
Laboratorieudvalget under Sygesikringens
og PLO´s Faglige Udvalg vedr. Almen
Praksis i samarbejde med DEKS og
Dansk Selskab for Klinisk Biokemi’s
Videnskabelige udvalg. Nov 2003.
Or www.SKUP.dk Kvalitetskrav til analyser
i almen praksis
2. Zwart A, van Assendelft OW, Bull BS,
England JM, Lewis SM, Zijlstra WG.
Recommendations for reference method for
haemoglobinometry in human blood
(ICSH standard 1995) and specifications for
international haemiglobinocyanide stan-
dard
(4th edition). J Clin Pathol. 1996 Apr;
49(4):271-4.
3. Method Comparison and bias estimation
using patient samples; approved Guideline.
Second edition 2000. NCCLS EP9-A2 vol
22. nr. 19
4. Mauro Buttarello, Quality specification in
haematology: the automated blood cell
count, Clin Chim Acta 346 (2004) 45 – 54
5. http://www.westgard.com/biodatabase1.
htm
6. Acceptability limits based on biology goals
in haematology EQAS. Skitek M. Accred
Qual Assur 2004; 10: 112-115
© 2005 THERMO ELECTRON CORPORATION. ALL RIGHTS RESERVED. ALL THE TRADEMARKS ARE THE PROPERTY OF THE THERMO ELECTRON CORPORATION AND ITS SUBSIDIARIES
Even the most accomplished clinical chemists and scientists need
the right tools and technology to quickly find definitive answers.
Thermo Electron provides high-throughput instruments and reliable
equipment that make analysis, diagnosis and conclusion possible.
From sample preparation and automation, to analyzers and
reagents, Thermo has the technology and experience to help
increase lab productivity, make diagnosis more accurate and
patients healthier.
For a closer look at all our clinical chemistry offerings, visit us at
www.thermo.com/konelab
Complete Clinical Chemistry
N
Fast and flexible TCAutomation systems for
automating preanalytical and analytical
phases
N
Fully automated clinical chemistry systems,
Konelab 20, 20XT, 30 and 60
N
High quality CE-marked Konelab diagnostic
kits for both routine and esoteric clinical
chemistry testing
Analyze
•
Detect
•
Measure
•
Control
TM
Look closer for answers
When you’re looking for answers others can’t see,
you need to look closer.
(Fortsat fra side 32)
