Indmedica Home | About Indmedica | Medical Jobs | Advertise On Indmedica
Search Indmedica Web
Indmedica - India's premier medical portal

Indian Journal for the Practising Doctor

HbA1c versus Mean Blood Glucose

Author(s): Verma A, Muthukrishnan J, Harikumar KVS, Modi K.D

Vol. 5, No. 5 (2008-11 - 2008-12)

ISSN: 0973-516X

HbA1c versus Mean Blood Glucose

Verma A, Muthukrishnan J, Harikumar KVS, Modi K.D

Dr KD Modi (Head of the Department), and Dr Abhyuday Verma , Dr J Muthukrishnan and Dr KVS Harikumar, (Senior Residents) , Department of Endocrinology, Medwin Hospitals, Hyderabad- 500001, India

Correspondence: Dr. Abhyuday Verma, Dept. of Endocrinology, Medwin Hospitals, Chirag Ali Lane, Nampally, Hyderabad- 500001, India. [Fascimile: +914066623441, Mobile: 09441132776 ; E-mail: abhyudaya76 (at) yahoo.com]

Abstract

HbA1c is an established parameter to assess mean glycemia over the preceding 3 months. Mean blood glucose value is derived from A1C report by a given formula based on DCCT data. However; A1C-derived average blood glucose overestimates blood glucose levels. Frequently patients with normal A1C report hypoglycaemia. Recently Nathan has suggested a correction in A1C-derived average blood glucose formula, which takes the mean value of multiple self-monitored blood glucose and, hence, is more representative of the actual mean blood glucose. The values of mean blood glucose derived from multiple blood glucose tests are getting more significance than only A1C-derived mean blood glucose. Value of this mean blood glucose value will be lower than the existing mean blood glucose value derived only from A1C level.

This simplification of the range will allow people with diabetes to understand their target level, particularly if already using home glucose monitoring. There is also a higher potential for future use as a diagnostic tool. Thus, Haemoglobin A1C levels might be replaced with mean blood glucose, a change that will add clarity for diabetic patients looking to manage their disease.

Key Words: glycated haemoglobin, mean blood glucose, self-monitoring blood glucose

HbA1c is an established parameter to assess mean glycemia over the preceding 3 months. Its correlation with diabetic complications shown in DCCT1 and UKPDS has made it a gold standard for therapeutic goals in diabetes. The mean blood glucose value is derived from A1C report by a given formula based on the DCCT data.3 However; A1C-derived average blood glucose overestimates the blood glucose levels. Frequently patients with normal A1C report hypoglycaemia and a normal A1C-derived mean blood glucose. Recently a study done by David M Nathan (Plenary session, ADA, June 2007), suggests correction in the ‘A1C-derived average blood glucose formula’. This new formula takes into consideration the mean value of multiple self-monitored blood glucose and hence is more representative of the actual mean blood glucose. With this formula the mean blood glucose values are found to be lower than the existing A1C-derived values from the DCCT data.

Increasing awareness requiring frequent self-monitoring blood glucose (SMBG) and availability of better, patient-friendly monitoring equipments that provide mean blood glucose measures, have helped patients reach their glycemic control better. However, no formula-derived parameter can be better than real time SMBG derived mean blood glucose, as reports are in the same units as the patients’ self monitoring of daily glycemia rather than as percent A1C. With new equipments, which provide mean blood glucose based on patients’ SMBG, this parameter can outdo HbA1c-derived average glucose as a future gold standard for assessment of glycemia. However, this will require large studies to establish its correlation with diabetes complications and its superiority over the existing gold standard HbA1c.

The glycohemoglobin assay (HbA1c/A1C) is the most widely used objective test of chronic glycemia in clinical and research applications. It is precise, reproducible and dependable. It is being used as a gold standard for monitoring glycemic control for over last two decades. It helps in guiding diabetes treatment and is a convincing tool to explain importance of glycemic control in prevention of major diabetic complications. The test was used as a marker of chronic hyperglycemia in major landmark clinical trials proving benefits of glycemic control in primary and secondary prevention of diabetic microvascular complications.1

The DCCT study done on Type 1 diabetic patients developed a normogram based on their frequently sampled blood testing.1 The mean blood glucose can be derived by specific statistical equation based on the HbA1c report (Mean blood glucose (mg/dL) = (30.9 × Hb A1C () – 60.6). This equation is derived from A1C reports of the DCCT trial.3 However A1C-derived mean blood glucose does not always correlate with the clinical situation. Many patients with apparently ideal A1C (5-6.5) and normal average blood glucose experience hypoglycaemia in real life. The advent of continuous glucose monitoring system and aggressive self-monitoring of blood glucose data have proved this.2 This was also observed in DCCT trial.15 On account of this correction in A1C derived mean blood glucose equation is suggested, based on multiple blood glucose monitoring results (ADA Plenary session June 2007).

The relationship between A1C and Plasma glucose (PG) is complex. On average, A1C of 6% corresponds to mean plasma glucose of 135 mg/dl. For every increase in A1C of 1%, mean plasma glucose increases by 35 mg/dl. Many studies have shown that A1C is an index of mean plasma glucose over the preceding weeks to months. A1C truly does not reflect glycemic control over last three months as it is claimed. Rather, it is weighted to the more recent weeks. The mean glycemia during the month preceding the A1C measurement contributes 50% of the result, during the 30-60 days prior to the A1C measurements contributes another 25%, and during the 60-120 days prior to the measurement contributes the final 25%.3

Fasting plasma glucose underestimates A1C (and seven-point mean plasma glucose) at increasing plasma glucose levels. On the other hand, post meal blood glucose contributes significantly to A1C. Post breakfast levels markedly overestimate A1C, whereas post lunch levels show a relationship to A1C that is very similar to that of mean plasma glucose.4

Calculating A1C has got its own disadvantages like; there are methodological and physiologic factors causing variation in A1C results. It does not reflect patterns of glycemia (when high and when low), the effects of individual foods or exercise, or the immediate response to changes in therapy. Physiologic factors that affect A1C include erythrocyte turnover rate. Any condition that shortens erythrocyte life span (such as a haemolytic anaemia or hypersplenism) will decrease A1C, conversely, aplastic anaemia, will result in aged circulating red blood cells and no new reticulocytes entering the pool, so A1C will progressively rise.5

Chemically modified haemoglobin, such as carbamylated haemoglobin associated with uraemia and acetylated haemoglobin formed after ingestion of large doses of salicylates, can falsely increase results.6 Kidney disease, liver disease, hemoglobinopathies, and recovery from blood loss will all decrease A1C. Vitamins C and E have been reported to lower A1C measurements, possibly by inhibiting glycation.7,8 Lower A1C levels are found in diabetic and nondiabetic pregnant women, probably due both to lower fasting blood glucose and a shortened erythrocyte lifespan.9 Iron-deficiency anaemia, has been associated with increased A1c.10 There is also some evidence of wide fluctuations in A1c between individuals that are unrelated to glycemic status, suggesting that there are “low glycators” and “high glycators”.11 Different laboratories and methods used yield different A1C values. For these reasons, no medical organization recommends the use of this test to diagnose diabetes.

To overcome these shortcomings of A1C calculations, recently International Federation of Clinical Chemistry (IFCC) has developed a new calibration standard that gives approximately 1.5% lower values. Existing A1C assay represents mixture of multiple glycated haemoglobins (young less glycated and old erythrocytes). Instead of that IFCC has suggested measurement of one standard, only A1C measurement and to be expressed in mmol / mol12. If implemented this may create great confusion for patients, clinicians and investigators.

The relationship between mean blood glucose and A1C is not always constant but differs depending on the glycemic control. Having lower mean glucose at the same A1C may help explain why intensive DCCT treatment had increased incidence of hypoglycaemia and decreased microvascular complications compared with conventional treatment.13 Earlier few studies tried to correlate A1C to average blood glucose (e.g. Svendsen et al 1982, Nathan et al 1983), suffered from relatively infrequent monitoring and small number of patients.

The DCCT had frequent A1c data in a large cohort (n=1441) of type 1 DM patients, but very infrequent capillary glucose data (e.g. stored data of 7 point capillary profile one day every 3 months). These infrequent glucose concentrations were not enough to compute a true ‘average’. Therefore despite our confidence in the ‘meaning’ of A1C assay according to DCCT the relationship between A1C and average glucose is not well established. The equation most commonly used, is according to DCCT data. (Table 1)

An alternate way to overcome this is to establish an exact relationship of the new results to mean blood glucose. With a new reference range, new targets, and a new name, reporting chronic glycemia in same units as the patients’ self monitoring of daily glycemia rather than as percent A1C, will be an advantage and will give a better understanding of glycemic control.14 Other advantages of mean blood glucose are that during DCCT it was a better predictor of the macro vascular complications of type 1 diabetes than HbA1c.15 A1C, mean blood glucose and glucose variability measurements each have an independent role in determining an individual’s risk of hypoglycaemia in type 1 diabetes.16

People have tried to derive a linear regression to convert A1C into average glucose. In order to do this a reliable regression model is required. An international study was needed to establish the relationship between A1c- average glucose across diabetes type, races, and ethnicities, where a confirmed mathematical relationship can be reported in the same unit as patients’ self monitoring of daily glycemia.

Recently a study, A1c Derived Average Glucose (ADAG), was done to analyse factors affecting A1c- average glucose relationship, to see that variability of blood glucose levels influences this relationship and to consider an alternative method to calculate average glucose.17 The preliminary results of this study were presented at plenary sessions of ADA held in June 2007. The results of this study showed a close correlation between A1C values at three months and average blood glucose during the same period. This relationship appeared to be the same for patients with type 1 and type 2 diabetes. New model for calculating average glycemia (Linear regression: average glucose = 29.573 × HbA1c – 52.346) and new glycemic values according to A1C levels were suggested. (Table1). The final results of this study will help in calculating average glucose from A1C more precisely.

Key message: HbA1c is a gold standard for assessing the mean glycemic control but the A1C-derived mean blood glucose overestimates blood glucose levels. Mean blood glucose derived from multiple, self monitoring blood glucose readings is more precise and helps in achieving a tighter glycemic control.

Conclusion

Overall, it appears that values of mean blood glucose derived from multiple blood glucose tests is getting more significance than only A1C derived mean blood glucose. Value of this mean blood glucose value will be lower than existing mean blood glucose value derived only from A1C level. This correlates well with actual clinical situation and explains hypoglycaemic episodes in patient with normal A1C and normal A1C derived mean blood glucose.

This simplification of the range will allow people with diabetes to understand their own target level, particularly if already using home glucose monitoring; and more likely would have a potential for future use as a diagnostic tool. Thus Haemoglobin A1C levels might be replaced with mean blood glucose, a change that will add clarity for diabetic patients looking to manage their disease.

References:

  1. Diabetes Control and Complications Trial (DCCT): results of feasibility study. The DCCT Research Group. Diabetes Care 1987;10:1-19.
  2. Christopher D Saudek, Rachel L Derr, Rita R Kalyani. Assessing glycemia in diabetes using self monitoring blood glucose and hemoglobin A1C. JAMA 2006; 295: 1688-1697.
  3. Beach KW. A theoretical model to predict the behavior of glycosylated hemoglobin levels. J Theor Biol 1979;81:547-61.
  4. Curt L Rohlfing, Jack D England, Hsiao-Mei Weidmeyer et al. Defining the relationship between Plasma Glucose and HbA1c. Diabetes Care 2002; 25: 275-278.
  5. Assessment of glycemia in Diabetes Mellitus: Hemoglobin A1c. JAPI 2005; 53: 299- 305.
  6. Chachou A, Randoux C, Millart H, Chanard J, Gillery P. Influence of in vivo hemoglobin carbamylation on HbA1c measurements by various methods. Clin Chem Lab Med. 2000;38:321-326.
  7. Davie SJ, Gould BJ, Yudkin JS. Effect of vitamin C on glycosylation of proteins. Diabetes. 1992;41:167- 173.
  8. Ceriello A, Giugliano D, Quatraro A, Donzella C, Dipalo G, Lefebvre PJ. Vitamin E reduction of protein glycosylation in diabetes. new prospect for prevention of diabetic complications? Diabetes Care. 1991; 14:68-72.
  9. Nielsen LR, Ekbom P, Damm P, et al. HbA1c levels are significantly lower in early and late pregnancy. Diabetes Care. 2004;27:1200-1201.
  10. Tarim O, Kucukerdogan A, Gunay U, Eralp O, Ercan I. Effects of iron deficiency anemia on hemoglobin A1c in type 1 diabetes mellitus. Pediatr Int. 1999; 41:357-362.
  11. Kilpatrick ES, Maylor PW, Keevil BG: Biological variation of glycated hemoglobin: implications for diabetes screening and monitoring. Diabetes Care 1998;21:261–264.
  12. Consensus Statement on the Worldwide Standardization of the Hemoglobin A1C Measurement: The American Diabetes Association, European Association for the Study of Diabetes, International Federation of Clinical Chemistry and Laboratory Medicine, and the International Diabetes Federation. Diabetes Care. 2007;30:2399–2400.
  13. Eric S. Kilpatrick, Alan S. Rigby, Stephen L. Atkin. Variability in the Relationship between Mean Plasma Glucose and HbA1c: Implications for the Assessment of Glycemic Control. Clin Chem. 2007 May;53(5):897-901.
  14. Sacks DB; ADA/EASD/IDF Working Group of the HbA1c Assay. Global harmonization of hemoglobin A1c. Clin Chem. 2005;51:681-683.
  15. Kilpatrick ES, Rigby AS, Atkin SL. Mean blood glucose compared with HbA(1c) in the prediction of cardiovascular disease in patients with type 1 diabetes. Diabetologia. 2007 Nov 27 [Epub ahead of print]
  16. E. S. Kilpatrick, A. S. Rigby, K. Goode, S. L. Atkin. Relating mean blood glucose and glucose variability to the risk of multiple episodes of hypoglycaemia in type 1 diabetes. Diabetologia. 2007 Dec;50(12):2553-61. Epub 2007 Sep 19.
  17. Shelley Wood. Hba1c Out, Average Glucose In? ADAG Results Support New Reference Method for Chronic Glycemia. Heartwire, Sept, 2007.

Legend:

Table 1: Existing and proposed A1C derived average blood glucose values

HBA1C Existing average
glucose according
to A1c mg/dl
(mmol/L)
PROPOSED A1C
DERIVED
Average Glucose
mg/dl (mmol/L)
5 100 (5.6) 96 (5.3)
6 135 (7.5) 125 (6.9)
7 170 (9.4) 153 (8.6)
8 205 (11.4) 184 (10.2)
9 240 (13.3) 214 (11.9)
10 275 (15.3) 243 (13.5)
11 310 (17.2) 273 (15.2)
12 345 (19.2) 303 (16.5)
Access free medical resources from Wiley-Blackwell now!

About Indmedica - Conditions of Usage - Advertise On Indmedica - Contact Us

Copyright © 2005 Indmedica