![]()
13
| 3 | 2008
Klinisk Biokemi i Norden
(Fortsætter side 14)
Methodological interferences
Earlier interferences like carbamylated Hb, the labile
Schiffs base or the ageing peak (HbA3 or P fraction)
are nowadays resolved and integrated in a proper
mode. The most common Hb variants worldwide like
HbS, HbC, Hb D and HbE are all seen in Scandinavia.
HbS and HbC are all resolved in the chromatograms.
A recent study investigated the effects of HbE and
HbD seen from the US horizon using 23 different
HbA1c assays (19). HbD originate from India but can
also be found in Sweden. HbE is the most common
variant in South East Asia and often combined with
beta-thalassemia trait. Due to the recent immigration
they are also seen in Scandinavia. HbE are underes-
timated in several HPLC methods. The combina-
tion with beta-thalassemia trait results in increased
turnover of erythrocytes resulting in lower HbA1c
not in agreement with actual blood glucose. More
than 25 Hb variants have been found during HbA1c
analysis by HPLC in Scandinavia and are characte-
rized. Approximately one third of them are related
to clinical dysfunction of the Hb molecule. Some
show increased oxygen affinity like Hb Malmö, Hb
Linköping (Finnish origin), others like Hb Köln, Hb
Leiden are related to haemolytic anaemia. Hb Tacoma
has been more frequently observed in Sweden with
new high resolution HPLC systems. Suspicion about
Hb abnormality must be related to clinical signs and
confirmed by MS-ESI (20) or DNA analysis. All chro-
matograms from HPLC equipment thus need careful
visual inspection.
Discussion
The quality goal for HbA1c varies between organi-
zations and geographical regions. Given the fact that
the risk of complication increases exponentially, it
has been estimated that to detect a change in risk of
10% would need a precision in the HbA1c assay with
CV below 1% and a negligible bias, which was judged
as unrealistic (21). It is also known that clinicians in
general bear unrealistic expectations of the quality of
HbA1c results (22).
The current quality goal for HbA1c among the
Swedish laboratories is a maximal bias of 0.2% (aver-
age deviation in EQA surveys over 6 month) and that
Foto: Ingunn Torsteinsdottir. Landmannalaugar, Island
