Printer Friendly

Unexpected hemoglobin a results after an erythrocyte exchange: importance of specimen mixing.

To the Editor:

Erythrocyte exchange, a procedure in which blood is removed and replaced with donor cells, is often used to prevent or treat severe vasocclusion in patients with sickling hemoglobinopathies. This procedure increases the percentage of hemoglobin A without dramatically increasing the hematocrit or viscosity. The efficacy of erythrocyte exchange is often measured by hemoglobin electrophoresis and densitometry to determine the posttransfusion percentages of hemoglobin A and S.

We performed a manual erythrocyte exchange on a 28-year-old pregnant woman with hemoglobin SD-Punjab. Five units of whole blood were withdrawn and replaced with packed erythrocytes and saline. The expected percentage of hemoglobin A in each unit withdrawn was calculated in an iterative fashion. We calculated the total volume of hemoglobin SD-Punjab erythrocytes using the patient's total blood volume estimated by nomogram, measured hematocrit, and assuming 100% hemoglobin SD-Punjab. The volume of erythrocytes removed was subtracted from the hemoglobin SD-Punjab erythrocyte volume tally, and the volume of erythrocytes administered (assumed to be an average of 160 mL) was added to the hemoglobin A erythrocyte volume tally. Volumes and percentages of hemoglobin A and SD Punjab erythrocytes in the patient and in the next volume of blood withdrawn could be estimated and the calculations repeated for each successive cycle. After the exchange, percentage hemoglobin A was measured by alkaline electrophoresis with densitometry, and a significant discrepancy between the calculated and measured value was noted. We report an investigation to elucidate whether the discrepancy was due to a methodological problem in the Sebia Hydrasys assay used at this institution.

We noted the manufacturer did not recommend a mixing step for sample preparation before withdrawal of erythrocytes from the washed cell pellet to prepare the hemolysate. Sickle erythrocytes have a heterogeneous density profile depending on the degree of dehydration, erythrocyte age, and other factors (1). In addition, transfused patients will have a dimorphic erythrocyte population of sickle cells and normal transfused cells. Given this heterogeneity, samples containing sickle cells may form a density gradient if allowed to settle. Depending on where the tip of the pipette lands in this gradient, different values for the percent of hemoglobin A could be obtained from the same patient specimen.


To test the theory that different values for the percentage of hemoglobin A could be obtained from the same patient specimen owing to a density gradient, we prepared 1 bag (bag 1) with the patient's pretransfusion blood and 4 bags (bags 2-5) containing increasing amounts of transfused blood and therefore increasing percentages of hemoglobin A. Two samples were taken from bags 2, 3, and 5, and hemolysate was prepared from each using 2 methods: the manufacturer's procedure for sample preparation and a variant in which a vortex-mixing step was added before withdrawal of the aliquot for hemolyzation. Also, 12 hemolysates were prepared from bag 4 (6 mixed and 6 unmixed) and analyzed in replicate.

Results for the hemoglobin electrophoreses of unmixed samples taken from the same bag yielded widely variable results (see Fig. 1). Not only was there a wider variation (range of SDs 2.9 -13.5 for unmixed vs 0.6 -2.2 for mixed), but the results were drastically different than those from the mixed sample from the same unit. The calculated hemoglobin A values were also closer to the results from the mixed samples. These findings suggest that thorough mixing before withdrawing erythrocyte aliquots during sample preparation distributes the cells more evenly and yields a more representative estimation of the percentage of hemoglobin A.

With a heterogenous erythrocyte population, the formation of a density gradient seems logical. However, current sample preparation methods do not address this potential source of error. According to the College of American Pathologists 2006 Survey (2), 58% of respondents use electrophoresis to measure the percentage of hemoglobin A. Three kits dominate: Sebia Hydrasys, Helena, and Beckman Coulter. In reviewing the manufacturers' sample preparation methods (3-5), we found that none included a vortex-mixing step during hemolysate preparation. However, we noted a recent change in the package insert for Sebia Hydrasys: the phrase "vortexmix them before taking 10 [micro]L to hemolyze" was added after instructions to discard the excess saline over the erythrocytes.

The problem brought to light by our investigation of the results for this single patient could represent a more pervasive issue, especially pertinent to those with sickling hemoglobinopathies. The omission of a vortex-mixing step seems not to be unique to a single kit or manufacturer but a common problem in hemoglobin electrophoresis sample preparation. Because the percentage of hemoglobinAafter an erythrocyte exchange is often used to guide treatment, recognition and correction of this omission are important.

Grant/Funding Support: None declared.

Financial Disclosures: None declared.


(1.) Rodgers GP, Schechter AN, Noguchi CT. Cell heterogeneity in sickle cell disease: Quantitation of the erythrocyte density profile. J Lab Clin Med 1985;106:30 -7.

(2.) College of American Pathologists 2006 participant survey for hemoglobin A electrophoresis. Northfield (IL): College of American Pathologists; 2006.

(3.) Hydragel 7 & 15 Hemoglobin(e) [package insert]. Norcross (GA): Sebia; 2006/09.

(4.) Helena procedural package inserts page. Helena Laboratories Website. Available at: Accessed March 28, 2008.

(5.) Paragon hemoglobin electrophoresis application guide EP-5 [package insert]. Fullerton (CA): Beckman Coulter.

Megan E. Sullivan (1,2)*

Anna Carolan (1)

James T. Perkins (1,2)

Irene J. Check (1,2)

(1) Evanston Northwestern Healthcare and

(2) Feinberg School of Medicine Northwestern University Evanston, IL

* Address correspondence to this author at:

Evanston Northwestern Healthcare

Department of Pathology and

Laboratory Medicine

2650 Ridge Avenue

Evanston, IL 60201

Fax 847-570-1938


DOI: 10.1373/clinchem.2007.100727
COPYRIGHT 2008 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2008 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Letters to the Editor
Author:Sullivan, Megan E.; Carolan, Anna; Perkins, James T.; Check, Irene J.
Publication:Clinical Chemistry
Article Type:Letter to the editor
Date:Jun 1, 2008
Previous Article:Improved HPLC analysis of serum 7[alpha]-hydroxycholest-4en-3-one, a marker of bile acid malabsorption.
Next Article:Insufficient standardization of a direct carbohydrateDeficient transferring immunoassay.

Related Articles
Hb and HCT do not match in oncology patients.
Effects of 7 hemoglobin variants on the measurement of glycohemoglobin by 14 analytical methods.
Human plasma ATP concentration.
Misleading glycated hemoglobin results in a patient with hemoglobin SC disease.
Quantitative evaluation of urinalysis test strips.
Effect of hemoglobin variants (Hb J, Hb G, and Hb E) on HbA1c values as measured by cation-exchange HPLC (diamat).
HPLC detection of fetal blood in meconium: improved sensitivity compared with qualitative methods.
Can glycohemoglobin be used to assess glycemic control in patients with chronic renal failure?
Carbamylated hemoglobin interference in Glycohemoglobin assays.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |