| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Department of Pathology, University of Utah, Health Sciences Center, Salt Lake City, UT and, ARUP Institute for Clinical & Experimental Pathology, 500 Chipeta Way, Salt Lake City, UT 84108, Fax 801-584-5207, e-mail william.roberts{at}aruplab.com.
To the Editor:
Hemoglobin (Hb) Constant Spring (CS) is a nondeletional form of 
-thalassemia that is most prevalent in southern Chinese and southeast Asian populations (1). The gene frequency ranges from 0.01 in central Thailand, 0.033 in northern Thailand, 0.05 in Laos, to 0.05–0.06 in northeastern Thailand (2)(3). Elongated chains of Hb are produced in low amounts and cause -thalassemia. In the homozygous state, the effects of this variant are more severe than the corresponding conditions that result from -globin gene deletion, and erythrocyte survival is markedly reduced(4).
Boronate affinity chromatography measurement of glycated Hb is generally considered resistant to interference by most Hb variants including HbC and S trait (5)(6). For this reason we have adopted a boronate affinity chromatography method as our routine method of glycohemoglobin measurement. Diabetic patient samples containing unusual variants are referred to our laboratory for glycated Hb testing when difficulties are encountered with cation exchange methods for HbA1c.
We describe a case of a sample from a patient with HbE trait and HbCS for which measurement of glycohemoglobin by boronate affinity chromatography was imprecise. Further investigation revealed that there was an extra peak in the chromatogram that interfered with proper integration of the glycohemoglobin peak (Fig. 1A
). This interfering peak was associated with the presence of HbCS in additional samples from patients without diabetes.
|
HbA1c was measured with an automated boronate affinity chromatography method (Primus Corporation). Although the method measures total glycated Hb, it is calibrated to report results as HbA1c equivalent values. Fructosamine was measured on a Modular P analyzer (Roche Diagnostics) with Roche reagents. All studies conducted using human samples were approved by the Institutional Review Board of the University of Utah.
The initial measurement gave a result of 8.9% HbA1c, but because a shoulder was observed on the glycohemoglobin peak, the sample was reanalyzed at the end of that same assay, giving a result of 7.0%. Based on these discrepant results, the sample was analyzed sequentially 10 additional times in the same assay. Results were 9.1%, 9.4%, 9.6%, 9.4%, 9.5%, 9.0%, 5.9%, 8.7%, 6.1%, and 8.9%, respectively. Inspection of the chromatograms revealed that for the results of 7.0%, 5.9%, and 6.1% peak integration was performed incorrectly due to the extra peak eluting between the nonglycated and glycated Hb peaks (Fig. 1A
). For samples with HbA1c results of 
9% the extra peak was integrated with the glycated Hb peak (Fig. 1B
). The overall CV for 12 replicates of this sample was 15.8%. Excluding the 3 lowest results gave a CV of 3.4%. A patient sample homozygous for HbA with an HbA1c of 8.2% was analyzed 10 times in 1 assay and gave a CV of 1.1%. The fructosamine concentration of a plasma sample from the index patient was 384 µmol/L (normal <285 µmol/L), consistent with an HbA1c result of 9%.
Phenotype analysis (HPLC, Variant analyzer, Bio-Rad Laboratories, beta-Thal short program) indicated 1.7% HbF, 4.6% P2 (HbA1c), 4.1% P3, 62% HbA0, 23% HbE, and 4.3% HbCS. HbCS eluted in 4 distinct peaks with retention times from 4.7–5.2 min. When we analyzed nondiabetic samples heterozygous for HbE (Fig. 1C
) or containing HbCS (Fig. 1D & E
), the sizes of the peaks eluting between nonglycated and glycated Hb were consistent with the concentration of HbCS in the samples. The presence of HbE did not produce such a peak.
The boronate affinity column that we used had a history of 3500 injections. Analysis of the sample from the index patients and other samples heterozygous for HbE and HbCS on a column from a different lot with <700 injections demonstrated no peak between nonglycated and glycated Hb. Additional testing of samples heterozygous for HbE and HbCS on this column after 3300 and 4400 injections failed to demonstrate a shoulder on the glycated Hb peak, suggesting that this interference from HbCS is specific to one lot of columns and mobile phase.
In conclusion, the presence of HbCS in samples for glycohemoglobin analysis by boronate affinity chromatography has the potential to interfere with accurate measurement of glycated Hb. Any boronate affinity chromatogram with a shoulder on the glycated Hb peak should lead to careful review of how the peaks were integrated and repeat testing by the same or a different glycated Hb method.
Acknowledgments
We gratefully acknowledge the excellent technical support provided by the staff of the Automated Endocrinology Laboratory and the Hemoglobin Identification section of the Special Genetics Laboratory at ARUP Laboratories.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
