![]()
32
| 2 | 2006
Klinisk Biokemi i Norden
33
| 2 | 2006
Klinisk Biokemi i Norden
(Fortsætter side 30)
evaluation are summarized in the table below.
For haemoglobin the quality demand stated in
Denmark for bias and imprecision are
≤
2% and
≤
3%, respectively. No agreed Scandinavian quality
goals exist for Leukocytes and 3 or 5 part differenti-
als. During the development of the protocol for this
evaluation, SKUP proposed the limits for total error
above, which should include
≥
95% of the results
for the quality goals to be fulfilled.
Haemoglobin:
The results achieved at the hospital
laboratory and in primary care did fulfil the quality
goals for bias, imprecision and total error in venous
samples.
In capillary samples the goals for total error less
than 5% was not fulfilled. In primary care there
was a positive bias of 2 – 2.6% (the demands were
≤
2%) combined with an imprecision of 3% (the
demands were
≤
3%). In hospital the bias was -1,7
% and imprecision 2,33 %. The results are similar
to previous SKUP evaluations for measurement
of haemoglobin in capillary blood (SKUP/2001/17
and SKUP/2004/29). The reason for the systematic
discrepancy between capillary and venous samples
is most likely preanalytical.
Leukocytes:
The results from the hospital laboratory
performed by an experienced laboratory technolo-
gist did fulfil the quality goals in both venous and
capillary samples for imprecision and total error.
The bias observed in the first primary care eva-
luation was reduced to 0 – 4.5% in the second
evaluation. The goals for capillary samples were
fulfilled in one of the two primary health care
centres. In the other centre only 83% of the capil-
lary samples were within ±16% of the result from
the comparison method. This was due to a positive
bias of about 4.5% combined with an imprecision
of 7.6%. In both primary health care centres the
precision was significantly better in venous samples
than in capillary samples, which might be attributed
to the capillary sampling technique.
The Three part differential
results did not fulfil the
quality goal suggested by SKUP for all the compo-
nents in both hospital and primary care.
For the
granulocytes
the goals were fulfilled in one
primary care centre and for capillary samples in the
hospital evaluation.
For the
lymphocytes
the goal was fulfilled in the
Results
Outpatients, hospital and primary care evaluation 2. Suggested quality goals with dark background
Total Error, percent fulfilling the goal
Range
Type of sample
Hospital
Two primary care
centres
Haemoglobin
(3 – 12 mmol/L)
± 5% deviation
Capillary
Venous
≥
83 %
≥
95 %
68 and 78 %
≥
95 %
Leukocytes
(3 – 25 x 109/L)
± 16% deviation
Capillary
Venous
≥
95 %
≥
95 %
83 % and 95%
≥
95 %
Granulocytes
(3 – 25 x 109/L)
± 23% deviation
Capillary
Venous
95 %
89%
87 and
≥
95 %
94 %
Lymphocytes
(0,3 – 4 x 109/L)
± 21% deviation
Capillary
Venous
95 %
92 %
85 % and
≥
95%
92 %
Monocytes
(0,4 – 2,5 x109/L)
± 43% deviation
Capillary
Venous
≥
95 %
≥
95 %
78 and 81 %
94 %
"EWARE OF THE MICROSCOPE EFFECT
7ORLD (EADQUARTERS s #ELLA6ISION !" )DEON 3CIENCE 0ARK 3%
,UND 3WEDEN s TEL
s INFO CELLAVISION COM
5 3 /FlCE s #ELLA6ISION )NC s
*UPITER 0ARK $RIVE 3UITE *UPITER &,
53! s TEL
s US INFO CELLAVISION COM
2EAD ABOUT THE #ELLA6ISION EFFECT AT WWW CELLAVISION COM
#ELLA6ISION© $- ANALYZERS AUTOMATICALLY LOCATE AND PRE
CLASSIFY THE BLOOD CELLS FOR YOU
5P TO SLIDES PER HOUR WITH AN IMAGE QUALITY AS GOOD AS IN A MICROSCOPE
(Fortsætter side 34)
(Fortsat fra side 31)
