Printer Friendly

Distribution of hemoglobin F, A, S, C, E, and D quantities in 4 million newborn screening specimens.

The California newborn screening program requires that all newborns be screened for selected hemoglobinopathies. Dried blood spot (DBS) specimens are analyzed using an automated 2-min cation-gradient HPLC method that is sensitive and specific (1, 2). The standardized method selected by the State allows the screening program to match the quantitative analytical performance of the assay at eight private contract laboratories and a central laboratory. Information technology is designed to derive phenotypes automatically and to use quantitative acceptability limits for quality control and proficiency testing.

The distribution of hemoglobin (Hb) quantities determined by cation-exchange HPLC in cord blood specimens has been published previously (3, 4). We have determined the frequency distributions for the analysis of DBS specimens using the rapid HPLC screening method. In this study, we examined screening data reported on 4 million nontransfused newborns tested within 2 days of birth. The frequency distributions were determined for the percentages of Hb concentrations in 14 phenotypes containing Hbs F, A, S, C, E, and D. The frequency distributions are available on request. The medians only are given in Table 1.

The percent concentrations in the current study are lower than the published cord blood data by 15% for all Hb F and by a median 36% for the Hb in other hetero- and homozygous patterns (range, 22-40%). This may be caused by differences in the HPLC methodology and by chemical degradation of Hb during formation of the dried blood spots (5, 6). Approximately 28-34% of the total area of the DBS chromatograms is eluted before the quantified species F, A, S, C, E, and D. These rapidly eluted species include unidentified degradation products and Fl, as well as a compound that elutes, in some samples, at the void volume and is correlated with Hb Barts (leftmost peak in the dashed curve in Fig. 1).

The relationships found among the concentrations of adult Hbs in carrier patterns are consistent with published cord blood data. As a fraction of the total adult Hb present in the newborn blood spots, the results are 0.41-0.45 for Hbs S, C, and D in patterns FAS, FAC, and FAD, and 0.27 for Hb E in FAE.

The chromatographic integration parameters have been adjusted to assure that all Hbs are reported in the pattern when the concentrations exceed 1% for Hbs F, A, E, and D and 0.5% for Hbs S and C (1). In the event that concentrations are below the limits, the software will continue to include the Hb in the pattern as long as the signal-to-noise criteria are satisfied (1). We have found that the one-percentile tails of the Hb frequency distributions are located above the specification limits in most cases. The one-percentile tail for Hb E in FAE, 0.7%, is not far below the detection limit of 1%, and we expect the Hb E to be reported accurately in the large majority of specimens. In the FSa patterns for S/[[beta].sup.+]-thalassemia, the median Hb S is 5.1% and Hb A is 0.9%. Thus, one-half the detected Hb A in FSa patterns is below the specification of 1% for the detection limit. For this reason, we expect that many cases of S/[[beta].sup.+]-thalassemia will be misclassified as FS. (Because all newborns with FS patterns are referred to follow-up, the differential diagnosis is made there.)


Typically, the shape of the frequency distributions and the relative standard deviations are approximated by a gaussian fit and are similar for the Hb concentrations in most patterns. However, anomalies were found in patterns FC and FAD. In FC patterns, the concentration of Hb C is unusually high at the 95th to 99th percentiles. In a span of 5 years, five of the eight specimens with the most Hb C were collected and tested at one clinical site. We have not been able to identify an analytical or demographic variable that would explain the high results, and more investigation is needed.

The frequency distributions for FAD show tails at the low end for Hb F and the high end for Hb A. These unexpected distributions are caused by nonspecificity of the HPLC method. True FAD patterns have a ratio of Hb D/Hb A similar to unity and always >0.5 (solid curve in Fig. 1). In the tails of the distributions for Hbs F and A, the chromatograms have a ratio <0.5. The low ratio is typical of an a-chain variant such as Hb G-Philadelphia, where the expected four chromatographic peaks are unresolved by the rapid screening method (7,8). One-half of the [alpha]-variant Hb F is eluted in the Hb A window, and the added two chromatographic peaks elute with Hbs A and D (Fig. 1). The two types of chromatograms are readily differentiated by calculating the ratio of the concentrations Hb D/Hb A.

Algorithms based on this ratio can be used to report the [alpha]-chain variants as such instead of as FAD patterns.

In summary, blood spot results using rapid cation-exchange HPLC give the following values. The median Hb [F.sub.0] is 67.5% in F-only patterns and 61.6% in all other patterns. (The concentration of Hb [F.sub.0] does not include the concentration of the acetylated Hb [F.sub.1].) The median Hb A is 10.3% in FA and approximately one-half of that value, or 5.3%, in FAS, FAC, and FAE. The expression of Hbs 5, C, and D in FAS, FAC, FAD, FSC, FSE, and FSD patterns is somewhat lower, at 4.0-4.2%, than the 5.3% Hb A. The Hb E is considerably lower, being only 2.0% in FAE and 1.6% in FSE. These results are consistent with published data that show that Hb variants have slightly lower concentrations than Hb A and that Hb E occurs at significantly lower concentrations than S, C, and D (3, 4). Because the co-inheritance of [alpha]-thalassemia affects the percentages of Hb found in carrier patterns such as FAS, clinical studies are needed to determine the frequency distributions for the different conditions.

In FS and FC patterns, the concentrations of Hb S and Hb C are 6.8%, which is less than twice the 4.0-4.2% expression in the heterozygotes and well below the 10.3% Hb A in normal FA patterns. The Hb E in FE is only 4.5%, which is less than one-half of the 10.3% Hb A in FA normals. We note that, in addition to homozygous hemoglobinopathies, the patterns FS, FC, FD, and FE reported by the rapid screening method include cases of [[beta].sup.0]-thalassemia and some cases of [[beta].sup.+]-thalassemia with Hb A below detection, as well as S with hereditary persistence of fetal Hb. Clinical studies are needed to determine the frequency distributions for the different conditions. One study has shown that in patterns FE, the phenotype EE can be differentiated in many cases from E/[beta]-thalassemia by the percentage of Hb E, which is 2.4-12.6% for EE and 1.4-4.5% for E/[beta]-thalassemia (9).


(1.) Eastman JW, Wong R, Liao CL, Morales DR. Automated HPLC screening of newborns for sickle cell anemia and other hemoglobinopathies. Clin Chem 1996;42:704-10.

(2.) Lorey F, Cunningham GC, Shafer F, Lubin B, Vichinsky E. Universal screening for hemoglobinopathies using high performance liquid chromatography: clinical results of 2.2 million screens. Eur J Hum Genet 1994;2:262-71.

(3.) Kutlar A, Kutlar F, Wilson JB, Headlee MG, Huisman THJ. Quantitation of hemoglobin components by high-performance cation-exchange liquid chromatography: its use in diagnosis and in the assessment of cellular distribution of hemoglobin variants. Am J Hematol 1984;17:39-53.

(4.) van der Dijs FPL, van den Verg GA, Schermer JG, Muskiet FD, Landman H, Muskiet FAJ. Screening cord blood for hemoglobinopathies and thalassemia by HPLC. Clin Chem 1992;38:1864-9.

(5.) Winter WP, Wozencraft LA. The chemistry of the decomposition of hemoglobin in filter paper blood samples used in hemoglobinopathy screening [Abstract]. Blood 1988;72(Suppl 1):214.

(6.) Roa PD, Turner EA, Del Pilar Aguinaga M. Hemoglobin variant detection from dried blood specimens by high performance liquid chromatography. Ann Clin Lab Sci 1993; 23:433-8.

(7.) Huisman THJ. Percentages of abnormal hemoglobins in adults with a heterozygosity for an [alpha]-chain and/or a [beta]-chain variant. Am J Hematol 1983;14:393-404.

(8.) Riou J, Godart C, Hurtrel D, Mathis M, Bimet C, Bardakdjian-Michau J, et al. Cation-exchange HPLC evaluated for presumptive identification of hemoglobin variants. Clin Chem 1997;43:34-9.

(9.) Lorey FW, Cunningham GC, Vichinsky E, Lubin B, Shafer F, Eastman J. Detection of HbE/[beta]-thalassemia versus homozygous EE using high-performance liquid chromatography results from newborns. Biochem Med Metab Biol 1993;49:67-73.

John W. Eastman, [1] * Fred Lorey, [2] John Arnopp, [2] Robert J. Currier, [2] John Sherwin, [1] and George Cunningham [2]

([1] California Department of Health Services, Genetic Disease Laboratory, 700 Heinz Ave., Suite 100, Berkeley, CA 94710, and [2] California Department of Health Services, Genetic Disease Branch, Berkeley, CA 94704; * author for correspondence: fax 510-540-2228, e-mail
Table 1. Median percentages of Hb concentrations in
newborn DBS specimens by rapid cation-exchange HPLC.

 V (a) = S, [F.sub.1]
Pattern [F.sub.0] A C, or D E + other

F only 67.5 32.5
Fsa (b) 61.6 0.9 5.1 32.4
FA 61.6 10.3 28.1
FAS, FAC, FAD 61.6 5.3 4.0 29.1
FAE 61.6 5.3 2.0 31.1
FSC, FSD 61.6 4.0 30.4
FSE 61.6 4.2 1.6 32.6
FS, FC, FD 61.6 6.8 31.6
FE 61.6 4.5 33.9

(a) V, variant.

(b) Only for cases in which the Hb A is detected.
COPYRIGHT 1999 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1999 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Technical Briefs
Author:Eastman, John W.; Lorey, Fred; Arnopp, John; Currier, Robert J.; Sherwin, John; Cunningham, George
Publication:Clinical Chemistry
Date:May 1, 1999
Previous Article:Different behaviors of serum and urinary crosslaps ELISA in the assessment of bone resorption in healthy girls.
Next Article:Altered composition of lipoproteins in liver cirrhosis compromises three homogeneous methods for HDL-cholesterol.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters