| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
|
|
|
|||||||
|
|||||||||
Electronic Letters to:
|
|
Electronic letters published:
![[Read eLetter]](/icons/misc/sectionDOWN.gif){kind=icon}
Incorrect Reference
HbA1c Determination In Patients With Uncommon Hb Variants
rebuttal
|
|
|||
|
Trefor Higgins, Co-Directors, Clinical Chemistry Dynacare Kasper Medical Laboratories, G Blakney
Send letter to journal:
thiggins{at}dkml.com Trefor Higgins, et al.
|
Drs. Bry, Chen and Sacks are to be congratulated on a well written, fully referenced article on the effects of hemoglobin variants and chemically modified derivatives on assays for glycohemoglobin. We would like, however, to point out that reference 45 which refers to an article written by us does not deal with the use of a 3 min elution program on the Tosoh A1C 2.2+ analyzer. We are not sure at which place in the text our article should be referenced. |
|||
|
|
|||
|
Wolfgang Schnedl, Karl Franzens University Graz, Austria Dept. of Internal Medicine, Auenbruggerpl. 15, A-8036 Graz, "Theresa Lahousen, Regina E Roller, Rainer W Lipp"
Send letter to journal:
wolfgang.schnedl{at}kfunigraz.ac.at Wolfgang Schnedl, et al.
|
To the Editor We read with interest the recent review by Bry et al. reporting on effects of hemoglobin (Hb) variants on assays for glycohemoglobin (1). A variety of hemoglobin variants are known to adversely affect the measurement of glycated hemoglobin (2). This review indicates that uncommon Hb variants that migrate with HbA1c produce spuriously increased HbA1c values upon ion-exchange chromatography. The authors state in table 1 that there is a Roche immunoassay that is unaffected by Hb Graz when HbA1c is measured (1). Our published findings with 4 immunological methods, including the Roche assays (Cobas Integra and Tina-Quant hemoglobinA1c tests, Roche, Vienna, Austria), showed that Hb Graz produced HbA1c values below the nondiabetic reference range in nondiabetic patients. In one type 2 diabetic patient with clearly diabetic, elevated blood glucose and fructosamine results, we demonstrated that HbA1c values determined with immunological methods were within the nondiabetic reference range (3). Only the HbA1c values obtained with the boronate affinity method (IMx glycated hemoglobin test, Abbot, Vienna, Austria) appeared to be in a clinically reasonable range. The review in question states that the Roche immunoassay is unaffected by Hb Sherwood Forest (1). Although an HbA1c determination method may produce a value in the normal range for a nondiabetic patient with a hemoglobin variant, this does not assure that no interference is present (3). It is conceivable that the interference is subtle in the normal range but increases with increasing HbA1c. In table 2 the variant HbD is listed as an uncommon Hb variant producing falsely low values for HbA1c (1). We described HbD as causing falsely low and high HbA1c values in ion exchange chromatography (4) and immunological methods (3). In general, ion exchange chromatography methods are not adequate for measurement of HbA1c in samples that contain hemoglobin variants, despite manufacturer’s claims to the contrary. An increasing number of Hb variants are recognized to falsify HbA1c results with immunoassays (2,3). Boronate affinity methods measure glycohemoglobin regardless of the glycation site and may be more useful in reflecting glycemic control in samples with Hb variants. Although this review is an important summary of the influence of various Hb variants in numerous HbA1c determination methods and the nature of these data is complex, the headings and information presented in tables 1 and 2 seem misleading. Additionally, in 3 referenced publications (references 17,18,22) there are 5 misspellings of names and one cited journal name is wrong (ref. 17 was published in Am J Clin Pathol 1995;104:444-6). References 1. Bry L, Chen PC, Sacks DB. Effects of hemoglobin variants and chemically modified derivates on assays for glycohemoglobin [Review]. Clin Chem 2001;47:153-63 2. Schnedl WJ, Liebminger A, Roller ER, Lipp RW, Krejs GJ. Hemoglobin variants and determination of glycated hemoglobin (HbA1c). Diabetes Metab Res Rev 2001;17:in press 3. Schnedl WJ, Krause R, Halwachs-Baumann G, Trinker M, Lipp RW, Krejs GJ. Evaluation of HbA1c determination methods in patients with hemoglobinopathies. Diabetes Care 2000;23:339-44 4. Schnedl WJ, Lipp RW, Trinker M, Hopmeier P. Hemoglobin D [beta 121(GH4)GluGln] causing falsely low and high HbA1c values in high performance liquid chromatography [Letter]. Clin Chem 1998;44:1999- 2000 |
|||
|
|
|||
|
lynn bry, clin chemist BWH and HMS, david sacks
Send letter to journal:
dsacks{at}rics.bwh.harvard.edu lynn bry, et al.
|
We thank the authors for their interest in our review and shall address their concerns. Concerning HbD, our review states on p. 157 that HbD can cause falsely low HbA1c values by certain chromatographic methods. The notion that Hb variants have complex effects on gHb measurements has been emphasized throughout the review, and we believe is one of the fundamental points of the article. With regard to the single, apparently non-diabetic patient with Hb Sherwood-Forest, the authors' data demonstrated normal HbA1c values by all non-HPLC HbA1c methods examined, including both immunoassay and affinity chromatographic methods (ref. 3 in the letter). These published results are accurately stated in Table I and in the text of the review. For the data in table I on Hb Graz, we specifically used only the non -diabetic information as the data concerning the single type 2 diabetic patient are confounding (ref. 3 in the letter). While the authors may have had data to indicate long-term poor control in this individual, only single fasting blood glucose (FBG) and fructosamine values were published with no additional clinical history. Although not stated, we assumed the analyses were performed concurrently. Fructosamine and gHb concentrations reflect the degree of glycemic control over the preceding ~2-3 weeks and ~2-3 months, respectively. Because the time frames of control measured by each analyte are not equivalent, we should have preferred a statement to the effect that fructosamine values were above the reference range for a period of three months, or some other objective indication of poor long- term glycemic control in this individual. Furthermore, while the fructosamine and FBG concentrations were well above the reference ranges, the gHb value by affinity chromatography was 6.7 (upper limit of reference range was 6.4). Given that affinity chromatography methods tend to demonstrate the least interference by Hb variants, a fact mentioned by the authors in their letter, we believe that the lack of indication of poor long-term metabolic control severely limits interpretation of the published laboratory data. The data concerning Hb Graz in table I of our review therefore focused on the measurements obtained from non-diabetic individuals. One individual had a value of 0.1 below the lower limit of the reference range (LLRR) with the Roche Tina-Quant assay, while the second was 0.3 below the LLRR. The results with the DCA-2000 methods were far more striking. For this reason, the DCA-2000 is not listed in Table I. However, given the DCA-2000 results, the review clearly stated on p. 157 that falsely decreased HbA1c values have been obtained with immunoassay methods on samples containing Hb Graz. We thank the authors for their comments and the issues raised. Interpretation of the literature on this subject is often confounded by many factors, including the limited numbers of patients with rare variants, and that studies often do not include the clinical information needed to fully assess adverse effects on gHb measurements. We agree with the authors that the effects of variant hemoglobins on gHb assays are often complex and can vary considerably among platforms and methods. |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
eLetters: