The relationship between blood glycosylated haemoglobin and home capillary blood glucose levels in diabetics

Diabetologia, Jul 1980

R. B. Paisey, D. G. Macfarlane, R. J. Sherriff, M. Hartog, R. R. Slade, D. A. J. White

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

The relationship between blood glycosylated haemoglobin and home capillary blood glucose levels in diabetics

Diabetologia R. B. Paisey ~ 0 D. G. Macfarlane t 0 R. J. Sherriff ~ 0 M. H a r t o g 0 R. R. Slade 0 D. A. J. White 0 0 University Department of Medicine, Bristol Royal Infirmary, and 2Department of Haematology, Southmead Hospital , Bristol, England 0 0 1 2 - 1 8 6 X / 8 0 / 0 0 1 9 / 0 0 3 1 / $ 0 1 . 0 0 Glycosylated haemoglobin; h o m e b l o o d glucose monitoring; diabetic control - Summary. Serial capillary b l o o d glucose levels f r o m insulin t r e a t e d patients w e r e r e c o r d e d o v e r 24 h o u r periods at fortnightly intervals for three months. T o t a l glycosylated h a e m o g l o b i n as % of H b A was m e a s u r e d at the end of this p e r i o d by the Fliickiger m e t h o d , and % H b A 1 by column c h r o m a t o g r a p h y . T h e r e were highly significant correlations b e t w e e n m e a n b l o o d glucose levels o v e r the three m o n t h s and % HbA1 (r = 0.93, 9 5 % confidence limits 0 . 8 4 - 0 . 9 8 ) , and with total glycosylated h a e m o g l o b i n (r = 0.88, 9 5 % confidence limits 0 . 7 5 - 0 . 9 4 ) . T h e r e was also a g o o d correlation b e t w e e n results o b t a i n e d by the two m e t h o d s (r = 0.81, p < 0.0001). T h e r e w e r e less strong correlations b e t w e e n % H b A 1 and b l o o d glucose levels during each of the three m o n t h s b e f o r e the estimation, with p e r c e n t a g e of glucose levels greater t h a n 10 mmol/1 and with m e a n fasting b l o o d glucose. T h e s e data support the hypothesis that % HbA~ and total glycosylated h a e m o g l o b i n are satisfactory m e a s u r e m e n t s of short t e r m diabetic control. In o r d e r to assess the relationship b e t w e e n hyperglyc a e m i a and diabetic complications, m e t h o d s of m e a s u r i n g glycaemic control o v e r long periods are highly desirable. H o m e m o n i t o r i n g of b l o o d glucose with reflectance m e t e r s has recently b e c o m e practicable [ 1, 2 ]. W e h a v e d e v e l o p e d an alternative filter p a p e r m e t h o d of capillary b l o o d spot collection b e c a u s e it is c h e a p e r and allows accurate l a b o r a t o r y b l o o d glucose assay [3]. F u r t h e r m o r e , m e a s u r e m e n t of the p e r c e n t a g e of glycosylated adult h a e m o g l o b i n ( % HbA~), [ 4-8 ], a p p e a r s to provide an integrated m e a s u r e of b l o o d glucose levels for a period of several weeks [ 9-12 ]. T h e present study was carried out to discover h o w closely glycosylated h a e m o g l o b i n (Hb), as m e a s u r e d by two different techniques, correlated with h o m e capillary b l o o d glucose levels t a k e n regularly o v e r the preceding three months. Patients and Methods Patients A group of cooperative and apparently fairly stable insulin dependent diabetics (Table 1) collected serial capillaryblood samples by finger prick on to filter paper strips. Most were treated with twice daily, mixed long and short acting insulins (total doses from 16-124 units daily); five "sulphonylurea failures" received Monotard 20-36 units each morning. Three patients had proteinuria but none had a raised serum creatinine level at the time of the study. They took samples seven to eight times during a 24 h period: i. e. before and 2 h after breakfast, lunch and supper, on going to bed, and sometimes between 02 00 and 04 00 h. The collection of the blood spots was demonstrated and patients were asked to take samples during a normal working day; informed consent was obtained from all. Patients who found the procedure acceptable and were able to produce adequate blood spots, were asked to take further 24 h series of capillary blood spots once every two weeks for three months. After this, blood was taken for measurement of Reason for insulin treatment Number Age years Failed sulphonylurea 10 treatment Ketosis prone dia- 29 betes from diagnosis Pregnant: ketosis 1 prone diabetes Age at % ideal diagnosisbody years weight 36-70 19-50 25 40-60 2-35 12 90-110 90-120 104 2B -SULPHONYLUREA FAILURES KETOSIS PRONE PATIENTS i P A T I E N T S ++~?247 ? ?,+ /+, + ?247 + + + ? n : 40 r = 0 . 9 3 p < 0.0001 I 2 i 4 I 6 I 8 I 10 I 12 i 14 I 16 Capillary blood glucose was measured in the filter paper spots by a modification [ 14 ] of the method previously described [ 3 ]. Filter paper previously soaked in 5% (w/v) boric acid was used. A 6 mm D 14 ,..J o (D a_ 12 U.I 5 O >- 8 m < "I" 6 0 diameter disc was punched from the centre of blood spots of adequate size and glucose eluted for one hour in 300 gl of 2V2% (v/v) sulphosalicylic acid. Glucose concentration in the eluate was measured on the glucose oxidase autoanalyser, bypassing the dialysis stage to increase sensitivity. Blood glucose was calculated on the basis that each disc of filter paper absorbed 11.0 _+ 0.4 bl blood (mean + 2 SD). Glycosylated Hb was measured by a colorimetric method [ 5 ], and % HbA 1 by "Quik-Sep" column chromatography. The latter is a commercial kit application of small column techniques to separate faster moving glycosylated haemoglobins (HbAl+a+b+c) from unaltered HbA [ 4, 6 ]. Our coefficient of variation for the Fliickiger method is 5.5% and for the column method 3%. The range of normal values, for the colorimetric and column assays are 5.3 _+ 1.6% and 7.2 _+ 1.2% respectively (mean + 2 SD). Results were analysed by least squares linear regression to obtain correlation coefficients. R e s u l t s Although the patients were selected for being cooperative and apparently well controlled, some showed considerable hyperglycaemia during the period of study (Fig. 1), and had considerable diurnal fluctuation of blood glucose. There were highly significant correlations between the measurements of glycosylated H b by both methods, and the mean of the blood glucose levels during the previous three months (Fig. 2), although the correlation was somewhat less close for the Fliickiger estimation (r = 0.88, p < 0.0001). Measurement of glycosylated H b on the same blood sample by the two methods showed good agreement (r = 0.88, p < 0.0001). The figures were further analysed to assess the degree of correlation of % H b A 1 with mean blood glucose levels during the first, second and third R. B. Paisey et al.: Glycosylated Haemoglobin and Home Blood Glucose Levels months before the estimation, the % of blood glucose levels > 10 mmol/1, the mean fasting blood glucose, and with the maximum fluctuations of blood glucose over 24 hours. The results (Table 2) showed that these indices of diabetic control did correlate with % HbA1, though none of the associations was as strong as that between the mean blood glucose during the whole of the previous three months and % H b A 1. Discussion Collection of blood samples on to filter paper strips is a satisfactory method for measuring blood glucose levels taken by patients at home. There were very close correlations between the mean 24 h blood glucose taken fortnightly during the previous three months and the levels of glycosylated haemoglobin particularly as measured by column chromatography. Others have also shown good correlations between blood glucose levels measured in out-patient clinics or in the ward, and the level of glycosylated haemoglobin [ 4-12 ]. On the whole, however, these correlations have been less close than those found in the present study. This supports the expectation that home blood glucose levels, taken during normal working days, are more representative of the usual prevailing levels of circulating glucose than in or outpatient measurements. Although the level of glycosylated haemoglobin in some way reflects the degree of hyperglycaemia in diabetic patients, the exact relationship between the two is still uncertain. Most studies to date have shown the closest correlations between % H b A 1 and the mean blood glucose levels throughout 24 hours in insulin treated diabetics [ 7, 10 ]. This was also our finding in our group of insulin treated diabetics. As in the study of Gonen et al. [10] we found that neither the peaks of hyperglycaemia nor the amplitude of fluctuation of blood sugar had a disproportionate effect upon blood HbA~ levels. As regards the period of time over which blood glucose levels affect the % HbA1, our best correlation was with mean blood glucose levels over the previous three months. Others have found correlations with blood glucose at the time of sampling [ 4 ], during the previous one to three months [ 6-10 ], and during the previous year [ 12 ]. One group found a correlation between % HbA1 separated by electrofocussing and urine testing in the previous eight weeks [ 9 ]. Most studies of blood HbA1 levels have involved the separation of the glycosylated haemoglobins by column chromatography. Column methods such as the one used in the present study are rapid and precise although the room temperature and p H of the eluting buffer are critical. The alternative colorimet~ ric method for total glycosylated H b correlated somewhat less well with the mean blood sugar than the column method, but might be preferred for routine use in some laboratories. Acknowledgements. We are indebted to the staff of the Departments of Chemical Pathology at the Bristol Royal Infirmary and Southmead Hospital for blood glucose determinations, to Mrs Lee Dowries and Mrs Gayle Barron for technical assistance, and to Miss Anne Brown for typing the manuscript. 1. S6nksen PH , Judd SL , Lowy C ( 1978 ) Home monitoring of blood glucose . Lancet I: 729 - 732 2. Walford S , Gale EAM , Allison SP , Tattersall RB ( 1978 ) Selfmonitoring of blood glucose . Lancet I: 732 - 735 3. Wakelin K , Goldie D J , Hartog M , Robinson AP ( 1978 ) Measurement of capillary blood glucose in filter-paper spots; an aid to the assessment of diabetic control . Br Med J II: 468 - 469 4. Trivelli LA , Ranney HM , Hong Tien L ( 1971 ) Haemoglobin components in patients with diabetes mellitus . N Engl J Meal 2 8 4 : 3 5 3 - 3 5 7 5. Fluckiger R , Berger W , Winterhalter KH ( 1977 ) Haemoglobin Alc, a reliable index of diabetic control . Diabetologia 13 : 393 6. Welch SG , Boucher BJ ( 1978 ) A rapid micro-scale method for the measurement of haemoglobin A 1 (a + b + c) . Diabetologia 14 : 209 - 211 7. Koenig RJ , Peterson CH , Jones RL , Saudek C , Lehrman M , Cerami A ( 1976 ) Correlation of glucose regulation and haemoglobin Alc in diabetes mellitus . N Engl J Med 295 : 417 - 420 8. Graf RJ , Halter JB , Porte D ( 1978 ) Glycosylated haemoglobin in normal subjects and subjects with maturity-onset diabetes . Diabetes 27 : 834 - 839 9. Lance R , Soria J , Thibault N , Soria C , Eschwege E , Tchobroutsky G ( 1977 ) Glycosylated haemoglobin concentrations and clinitest results in insulin-dependentdiabetes . Lancet II: 1156 - 1157 10. Gonen B , Rubenstein AH , Rochman H , Tanega SP , Horwitz DE ( 1977 ) Haemoglobin A~: An indicator of the metabolic control of diabetic patients . Lancet II: 734 - 736 11. Bunn HF , Gabbay KH , Gallop PM ( 1978 ) The glycosylation of Hb; relevance to diabetes mellitus . Science 200 : 21 12. Elkeles RS , Wu J , Harnbley J ( 1978 ) Haemoglobin A1, blood glucose and high density lipoprotein cholesterol in insulinrequiring diabetics . Lancet II: 547 - 548 13. Hill JB , Palmer PP ( 1969 ) Filter paper blood collection and punching as a means of quantification . Clin Chem 15 : 381 14. Paisey R , Bradshaw P , Hartog M , West P ( 1979 ) Home monitoringof blood glucose using filter paper strips . Br Med J II: 1509 Received: October 9 , 1979 , and in revised form: March 7 , 1980 Dr. R. B . Paisey Department of Medicine Old Building Bristol Royal Infirmary Bristol BS2 8HW England

This is a preview of a remote PDF:

R. B. Paisey, D. G. Macfarlane, R. J. Sherriff, M. Hartog, R. R. Slade, D. A. J. White. The relationship between blood glycosylated haemoglobin and home capillary blood glucose levels in diabetics, Diabetologia, 1980, 31-34, DOI: 10.1007/BF00258307