ANNALS O F CLINICAL AND LABORATORY SCIEN CE, Vol. 14, No. 4 Copyright 1984, Institute for Clinical Science, Inc. Influence of Methodology on Glycosylated Hemoglobin Values in Nigerian Subjects with Sickle Cell Hemoglobinopathy A. E. OHWOVORIOLE, M.B., F.M.C.P., J. A. KUTI, M.B., M.R.C.P., andt. O. JOHNSON, M.D., F.R.C.P. Endocrine and Metabolic Unit, D epartm ent o f Medicine, Lagos University Teaching Hospital, P. M. B. 12003, Lagos, Nigeria ABSTRACT The determination of glycosylated hemoglobin (GHb) has been found useful in the medium term assessment of diabetic control. Levels of GHb have been shown to be influenced by hematological abnormalities and methodology amongst other factors. Hemoglobinopathies are common in Nigerians and other people of African descent. No work, to our knowledge, has been done in Nigeria to study the effect of hemoglobinopathies on the levels of GHb in Nigerians or Africans. In this study, GHb levels in non-diabetic Nigerians (with hemoglobin genotypes AA, AS, and SS) were determined by short column chromatography and thiobarbituric acid colorimetry. Levels of GHb obtained by the microchromatographic method were significantly different from one another in the three subject groups. The group mean GHb level was highest in subjects with sickle cell anaemia and lowest in subjects with sickle cell trait. However, using the colorimetric method, the mean GHb level of the normal subjects with hemoglobin genotype AA was not significantly different from that of subjects with hemoglobin genotype AS or SS. Results of GHb determinations by microchromatography (for which several commercial kits are available) in subjects with sickle cell anemia or trait must be interpreted with caution. The colorimetric method, though more tedious, gives more reliable results and should be the method of choice in subjects with concurrent diabetes and sickle cell anemia or trait. Introduction The minor glycosylated hemoglobins have gained renewed attention following the observation that this component of the normal hemoglobin is increased in diabetics (II). Determination of the level of glycosylated hemoglobin (GHb) could be used to m onitor diabetic control. Levels of GHb provide an integrated 265 0091-7370/84/0700-0265 $00.90 Institute for Clinical Science, Inc.
2 6 6 OHWOVORIOLE, KUTI, AND JOHNSON index of the mean blood glucose concentration during the preceding several weeks.3 There are now several methods available for the determination of GHb, viz., short and long-column cation resin chromatography, high perform ance liquid chromatography, iso-electric focusing, and the colorimetric thiobarbituric acid (TBA) methods. The column resin chromatographic and the TBA methods are the ones most widely used in clinical practice. The column chromatographic method measures all three components of GHb (HbAxa + b + c) and the TBA colorim etric m ethod m easures HbA l c which accounts for about 80 percent of total GHb levels. Using the shortcolumn chromatographic method, results can now be obtained within a few hours of sample collection.4 The colorimetric method takes several hours to complete and is generally more tedious. However, the chromatographic method is influenced to a great extent by storage conditions of the blood, ambient temperature, anem ia and hem oglobinopathies amongst other factors.5 Hemoglobinopathies which are common in Africans and peoples of African descent could thus modify the interpretation of GHb levels in the management of African diabetics. In this study, the effect of sickle cell trait (AS) and sickle cell anaem ia (SS) was determ ined on the levels of GHb as measured by the two common m ethods microchromatography and colorimetry in a tropical setting. Subjects, Materials and Methods S u b j e c t s Three groups of non-diabetic Nigerians were studied. The characteristics of the subjects are shown in table I. The m ean age of the subjects with sickle cell anemia (SCA) was significantly TABLE I Characteristics of Subjects Studied Number of Male: Mean ± SD Subjects Female (years) Mean RBG Hemoglobin AA 33 20:13 25 ± 3.5 80.0 ± 0.9 Hemoglobin AS 12 8.4 24.0 ± 4.2 77.3 ± 11.8 Hemoglobin SS 20 8.12 18.4 ± 2.5 72.4 ± 10.6 Diabetics Ì 15.12 43.5 ± 10.2 175.6 ± 37.2 *Random blood glucose in mg per dl. lower than that of the subjects with hemoglobin genotypes AA and AS (P < 0.05). The diabetics were on various regimes of treatment. Non-diabetic subjects were recruited from medical staff and students as well as patients who had no evidence of carbohydrate intolerance. Subjects with SCA were recruited from the sickle cell clinic of the Lagos University Teaching Hospital, Lagos. A subject was regarded as non-diabetic if the fasting and/or twohour postprandial blood glucose level was less than 120 mg per 100 ml and 140 mg per 100 ml, respectively, modified from W H O.13 Materials and Methods Venous blood for glycosylated hemoglobin (GHb) was obtained from all subjects into EDTA tubes. Total GHb was determined by microchromatography using Biorad kits as described previously.10 Results are expressed as percent GHb (HbALpercent). H em oglobin Axc (HbAxc) was d e te r mined by the TBA colorimetric method employing a modification of the method described by Fluckiger and W interhalter6 and modified by Gabbay et al.7 About 3.0 ml of whole blood in EDTA was washed three times with about seven ml norm al saline by spinning for five minutes and aspirating the supernatant to waste. After the last wash, about one ml of distilled water was added to the red
HEMOGLOBIN VALUES IN SICKLE CELL HEMOGLOBINOPATHY 2 6 7 cells and mixed. To the hemolysate, an equal volume of chloroform was added, vortex-mixed for three minutes, and then spun for 20 minutes. Hemolysate so prepared was adjusted with distilled water to give a hemoglobin concentration of 10 g per liter. To 1.0 ml of the hemolysate, 0.5 ml of 0.5 molar solution of oxalic acid was added and vortex-mixed. Fructose standards were prepared to give concentrations of 0, 0.36, 0.72, 1.09, 1.44, and 2.16 mg per 100 ml. After mixing, the tubes w ere then covered with ru bber stoppers vented with 26 gauge hypodermic needles. The needles were removed after about 10 min when equilibration m ust have occurred; heating continued for five hours in a water bath. The tubes were then allowed to cool in an ice bath. Cold 40 percent trichloroacetic acid 0.5 ml was added to each tube, mixed, and centrifuged at 3000 rpm for five minutes. To 1.5 ml of the supernatant was added 0.5 ml of 0.05 mol per titre of TBA and it was then mixed. After incubating at 40 C for 40 min, the tubes were allowed to cool at room tem perature and then read at a wavelength of 443 nm in a Unicam SP 500 spectrophotometer. A fructose standard curve permits the conversion of absorbances of hem olysate preparations to fructose equivalents, i.e., mg fructose per 10 mg hem olysate. Alternatively, the HbAjC values can be expressed as absorbance at A443 per g of Hb or A443 per 10 mg of hemolysate. Both units have been found to correlate well with total GHb, using more precise methods.712 Expressing the results as A443 per unit of hemoglobin simplifies the procedure further and renders the fructose standards unnecessary. Hemoglobin electrophoresis was performed on cellulose acetate paper as described.9 Of the 20 subjects with SC A, hemoglobin F (HbF) band was observed in 12. No subject with HbAS or Hb AA showed HbF band. Subjects with hemoglobinothies like HbAC or HbCC were not included in the study. Statistical analysis is by the student s t test for paired and unpaired data, where appropriate. The level of statistical significance is taken as P < 0.05. Results P r e c is io n Using ten replicate samples from a pooled hemolysate on non-diabetics with hemoglobin genotype AA, the intra-assay coefficients of variation (CVs) were 7.5 percent and 3.6 percent, respectively, for the microchromatographic and colorimetric methods. Interassay CVs from ten determinations of a pooled hemolysate were 10.9 percent and 6.1 percent, respectively, for the chrom atographic and colorimetric methods. T o t a l G l y c o s y l a t e d H e m o g l o b i n a n d H ba jc V a l u e s In table II are shown the values of total GHb (HbAj a + b + c) of the various subject groups as determined by microchromatography. The group mean HbAj values of the non-diabetic subjects with hemoglobin genotypes AS and SS are significantly different from the group mean HbAj value of subjects with hemoglobin TABLE I I Total Glycosylated Haemoglobin Levels By Microchromatography S i g n i f i c a n c e T o t a l % HbA^ D i f f e r e n c e M ean ± SEM fr o m N o r m a ls Number o f S u b j e c t s (R a n g e ) (G e n o ty p e AA) Normals 33 7.63 + 0.35. (5-9) Sickle cell trait 12 6.50 + 0.43 P < 0.05 (4-9) Sickle cell anemia 20 9.73 + 0.38 P < 0.01 (7.5-14.0) Diabetics 27 11.51 + 0.56 P < 0.01 (6.5 15)
26 8 OHWOVORIOLE, KUTI, AND JOHNSON genotype AA. The mean (SD) HbA: 11.7 (1.62 percent) of SCA subjects with HbF bands was significantly higher than the mean HbAx 8.81 (1.06 percent) of SCA subjects without HbF bands on electrophoresis (P < 0.001). In table III are shown the H ba ^ values obtained by the TBA colorimetric method, and expressed in two ways, mg F per 10 mg Hb and A443 per 10 mg of hemolysate. The means of the various non-diabetic subject groups are not significantly different from one another. The mean of the diabetics is significantly higher than any of the non-diabetic groups (P < 0.01). Among the SCA subjects, there was no significant difference between the mean (SD) values of subjects with HbF bands and those without, 0.160 (0.14) and 0.169 (0.016), respectively. C o r r e l a t i o n s In subjects w ith hem oglobin genotypes AA and AS, the coefficients of correlation betw een chromatographic and colorimetric values were r = +0.852 and 0.709, respectively. Correlation between colorimetric and microchromatographic values was weak in subjects with sickle cell anemia (r = +0.187). There was a high correlation between results of HbA,C expressed as mg of fruc- TABLE I I I HbAjc Levels by Thiobarbituric Acid Colorimetry HbAj_c HbAjC as mg Fructose/ as A443/ Number of Subjects ^ msr Hb 10 mg Hb Normal 33 0.6 8 + 0.019 0.1 70 + 0.003 (0.53-0.84) (0.144-0.200) S ic k le c e l l t r a i t 12 0.6 4 + 0.039* 0.172 + 0.004* (0.46-0.87) (0.143-0.195) S ick le c e ll anemia 20 0.61 ± 0.036* 0.163 + 0.004* (0.40-0.81) (0.164-0.405) D iabetics 27 1.43 + O.lOf 0.268 ± O.lOf (0.53 2.93) (0.164 0.405) R esu lts expressed as mean ± SEM ( ran g e ). Variation from norm al v a lu e s n o t s i g n i f i c a n t, f s i g n i f i c a n t v a r i a t i o n from norm al v a lu e s (P < 0.0 1 ). tose per 10 mg hemoglobin and A443 per 10 mg of hemoglobin (r = +0.988, P < 0.01). Discussion Of the two common methods for determ ining GHb levels, the microchromatographic method which is available in com m ercial kits is much faster and more convenient. The chromatographic method is, however, subject to various influences including environmental temperature and hematological abnormalities. The results of this study show that the effect of sickle cell hemoglobinopaties on GHb levels as determined by microchromatography could be considerable. The results of the TBA colorimetric method, on the other hand, are little affected by sickle cell anemia or trait as shown in this study. These results are in agreement with the yet small number of other studies in American Blacks and South Africans.1,212 The poorer correlation between values obtained by the two methods described in this paper in subjects with sickle cell anemia or trait is due to the behavior of the hemoglobin variants during chromatography. In column chromatography of hemolysate, glycosylated variant hemoglobins like HbS and HbC that are more positively charged than HbA: migrate with the norm al HbA and, therefore, total GHb levels are falsely low. On the other hand, HbF migrates with the fast HbAj a + b. Therefore, in individuals with persistent H bf a common finding in subjects with hemoglobinopathies total GHb levels determined chromatographically may be falsely elevated. To be employed usefully, GHb values obtained by the chromatographic method in diabetic subjects with sickle cell trait or anemia have to be adjusted, taking into account the amount of abnormal hemoglobin present.12 Derivation of such correction factors requires additional equipment for the quantitation of the ab
HEMOGLOBIN VALUES IN SICKLE CELL HEMOGLOBINOPATHY 2 6 9 normal hemoglobin. Facilities for such quantitation are not readily available in the average tropical hospital set-up. The colorimetric method measures the stable fraction of the subfraction HbA1c, i.e., the glucose bound in ketoam ine linkage to the hemoglobin molecule.7 It is, therefore, not affected by the factors that influence results obtained chromatographically. In addition, this rather tedious and long method is more reproducible, cheaper, and requires little additional equipment of the hospital laboratory. This w ould appear to be the method of choice for the average tropical hospital laboratory. In diabetics with abnormal hemoglobins, determ ination of GHb levels in the monitoring of diabetic control should preferably be by colorimetry rather than by microchromatography. Acknowledgments Thanks are extended to M essrs. J. O lorundu and T. Akpan and to D rs. S. O. Salami and C. Chukwu for their assistance, and to Professor O. O. Akinyanju for allowing us to study his patients w ith sickle cell anem ia. W e thank Mr. T. M. Adade for secretarial help. The paper was p resen ted in part at th e 11th ID F Congress, Nairobi, 1982. References 1. A l e y a s s i n e, H. : L ow proportions o f glycosylated hem oglobin associated with hem oglobin S and h em o g lo b in C. C lin. C h em. 25:1484 1486, 1978. 2. B e r n s t e i n, R. E.: Haematologic considerations d eterm ine w hich assay for glycohem oproteins is advisable (L etter). Clin. C hem. 2 6 :1 74-175, 1980. 3. B u n n, H. F.: N on-enzym atic glycosylation of proteins: R elevance to diabetes. Am. J. M ed. 70:3375-3380, 1981. 4. C o l e, R. A., S o e l d n e r, J. S., D u n n, P. J. and B u n n, H. F.: A rapid m ethod for the d ete r m ination of glycosylated hem oglobins using high p ressu re liquid chrom atography. M etabolism 27:289-301, 1978. 5. E d i t o r i a l : H aem oglobin A t and diabetes: A reappraisal. Brit. M ed. J. 281:1304-1305, 1980. 6. F l u c k i g e r, R. and W i n t e r h a l t e r, K. H.: In vitro synthesis of hemoglobin AjC. F E B S Letters 77:356 360, 1978. 7. G a b b a y, K. H., S o s e n k o, J. M. B a n u c h i, G. A., M i n i n s o h n, M. J., and F l u c k i g e r, R.: Glycosylated hem oglobins: Increased glycosylation of hem oglobin A in d iabetic patients. D iabetes 28:337-340, 1979. 8. G o l d s t e i n D. E., P a r k e r, K. M., E n g l a n d, S. D., E n g l a n d, J. E., W i e d m e y e r, N., R a w l i n g s, S. S., H e s s, R., L i t t l e, R. R., S i m m u n d s, S. F., and B r e y f o c l e, R. P.: Clinical application of glycosylated hemoglobin m easurements. D i abetes 31 (suppl 3):70-78, 1982. 9. H e l e n a E l e c t r o p h o r e s i s M a n u a l (1975). H elena Laboratories, Beaumont, TX 77704. 10. K u t i, J. A. O h w o v o r i o l e, A. E. and J o h n s o n, T. O. : HbA[ levels in Nigerians. Nig. Q uart. J. Hosp. Med. i:2 1-2 3., 1982. 11. R h a b a r, S.: Abnormal hem oglobin in red cells of diabetics. Clin. Chim. Acta 22:296 298, 1968. 12. S o s e n k o, J. M. F l u c k i g e r, R. L., P l a t t, O. S., and G a bbay, K. H.: Glycosylation of variant hemoglobins in normal and diabetic subjects. Diabetes Care 3:590 593, 1980. 13. W.H.O.: Expert Com m ittee on Diabetes M ellitus, 2nd Report. Technical R eport Series No. 646, 1980, p. 10.