| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Department of Clinical Biochemistry, School of Technology for Medical Science, and
2
Department of Public Health, School of Medicine, Kaohsiung Medical University, Kaohsiung 80702, Taiwan
a author for
correspondence: fax 886-7-2370544, e-mail
tsliyu{at}cc.kmu.edu.tw
Hemoglobin A1c (HbA1c) is used for
the long-term management of patients with diabetes mellitus (DM)
(1)(2). Hb variants other than HbA1c and
-N-lysine-glycated Hb A0 may cause analytical
interference in determinations of HbA1c (3)(4)(5)(6). In one
study, the authors estimated the prevalence of thalassemia in Taiwan as
7%; moreover, 
1% of the people in northern Taiwan are
ß-thalassemia heterozygotes (7). The occurrence of 24
abnormal Hbs (13 
-chain variants and 11 ß-chain variants),
including Hb G-Taipei, in populations in the Silk Road area of
Northwestern China has been presented in a review (8). The
frequency of thalassemia has been estimated to be 1 in 2350 in Japan
(9) and even higher in North Africa (10). Hb E is
the second most prevalent Hb variant worldwide and the third most
prevalent variant in the US, after Hb S and C. Hb E is found primarily
in Southeast Asia, especially among the Thai population
(11). In the northeastern region of India, the gene
frequency of Hb E is 10.9% (12). In a study of 222 000
blood samples in Canada, 23 cases of Hb J were identified
(13). Given that the majority of hemoglobinopathic cases are
from families of Asian, Southeast Asian, and Asian Indian ancestry
(7)(8)(9)(10)(11)(12)(14)(15)(16), the aim of this study was to
investigate the influence of selected Hb structural variants on HbA1c
values measured by cation-exchange HPLC.
We collected 17 EDTA-anticoagulated whole blood specimens from DM patients with Hb AJ (6 patients), Hb AG (10 patients), or Hb AE (1 patient had a fasting sugar of 10.4 mmol/L) to analyze HbA1c. The ranges and mean values for fasting sugar were 8.2–17.8 mmol/L and 12.6 ± 3.9 mmol/L, respectively, in the DM patients with Hb AJ and 7.1–18.1 mmol/L and 9.8 ± 4.1 mmol/L, respectively, in the DM patients with Hb AG. In addition, one specimen from a nondiabetic patient with the Hb AG variant (fasting sugar, 4.4 mmol/L) and another from a nondiabetic patient with Hb AE (fasting sugar, 5.2 mmol/L) were analyzed. HbA1c and glycated Hb were measured by cation-exchange HPLC (Diamat HbA1c program; Bio-Rad Laboratories) and by boronate ion capture (IMx analyzer; Abbott Laboratories). Both methods had a CV <5%, and both reported results as percentage of HbA1c. When Tiran et al. (4) comparatively evaluated five glycated Hb assay methods, including the Abbott IMx glycated Hb ion capture assay, they found that the methods showed generally acceptable precision and good accordance with the Bio-Rad Diamat system. Bon et al. (17) determined the accuracy of the IMx assay by comparison with a reference HPLC assay for 603 specimens; the correlation coefficients were 0.88–0.96. In addition, several investigators have shown that glucose, bilirubin, triglycerides, labile fraction, and Hb variants do not interfere with the Abbott IMx assay (18). Moreover, the IMx assay is not sensitive to interference by cyanate derived from spontaneous dissociation of urea. In the present study, a boronate-affinity analytical method on a CLC 385 analyzer (Primus Corporation) served as the comparison method because of the high specificity and the negligible interference of Hb variants in that method (1). The Hb variants were identified by electrophoretic separation of Hb on cellulose acetate membranes. Specimens for which HPLC chromatograms suggested the presence of abnormal peaks underwent hemoglobinopathy studies.
The Abbott IMx boronate ion-capture method showed no important effects
from any of the Hb variants tested, and its results for HbA1c agreed
well with those of the comparison method
(r2 = 0.93, n = 43 for HbA1c;
r2 = 0.95, n = 43 for glycated
Hb), whereas the results analyzed by cation-exchange HPLC
method were falsely low (Fig. 1A
). Three different chromatographic patterns were
observed in blood specimens with HB J (6 samples), Hb E (2 samples),
and Hb G (11 samples). An asymmetrical HbA1c peak with a "left
shoulder" appeared in the specimens with Hb AJ (Fig. 1B
, chromatogram
A), and an asymmetrical HbA1c peak with a "right shoulder" appeared
in the specimens with Hb AE (Fig. 1B
, chromatogram B). For specimens
with Hb AG, the HPLC chromatogram showed an additional peak at Hb
A0 (Fig. 1B
, chromatogram C). HbA1c values
measured by affinity chromatography were appropriately increased for
the patients’ blood glucose values, but HbA1c values measured by HPLC
were lower than those measured by affinity chromatography, whether the
patients were diabetic or not. Thus, the Hb mutations studied caused an
abnormal HPLC chromatogram and falsely low HbA1c values when measured
by HPLC.
|
The effects of Hb variants on HbA1c values determined by the HPLC system have been evaluated previously (3)(5)(6)(19)(20). An earlier report by Oshima et al. (5) of a study conducted in Japan described an abnormal chromatogram for HbA1c [as analyzed on an automated glycohemoglobin analyzer, HLC-723 Ghb V (Tosoh)] from a male DM patient with Hb J Lome. In a study performed in Singapore, Wong et al. (19) observed an asymmetrical HbA1c peak with a right shoulder in nine HbA1c blood specimens analyzed by cation-exchange HPLC (Variant HbA1c program; Bio-Rad) from diabetic patients with Hb AE. In a study performed in Austria, Schnedl et al. (20) reported an additional peak at Hb A0, as well as falsely low HbA1c values, measured by cation-exchange HPLC (Diamat HbA1c) from a diabetic patient with Hb O Padova. Our results not only agree with those of Schnedl et al. (20), but also deal with Hb J, G, and E variants in a single study.
We conclude that Hb variants can contribute to mismanagement of patients with DM because of falsely low HbA1c values measured by HPLC. Careful interpretation of glycohemoglobin results is critical in populations with a relatively high prevalence of Hb variants, such as Hb AJ, AG, and AE.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
