The Effects of Aging on Hemoglobin A1c Levels—The Potential Role of the Glycemic Gap

Abstract

The diagnosis of diabetes has come a long way since its initial description 3,000 years ago by the Egyptians.¹ Until the 19^th Century, the sweet taste of glucose in the urine was used for diabetes diagnosis; chemical methods became available to measure blood glucose around 100 years ago.²

In 1979, the National Diabetes Data Group provided diagnostic criteria for diagnosing diabetes that relied on distributions of glucose levels, rather than on the relationship of glucose levels with complications, despite emerging evidence that the microvascular complications of diabetes were associated with a higher range of fasting and oral glucose tolerance test (OGTT) glucose values.³

In 1997, the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus refocused attention on the relationship between glucose levels and the presence of retinopathy as the basis for the diagnosis of diabetes. They also lowered the fasting plasma glucose (FPG) cut point to ≥126 mg/dL (from 140 mg/dL), so that this cut point would represent a degree of hyperglycemia that was “similar” to the 2-h plasma glucose (PG) by OGTT value of 200 mg/dL, and diagnosis with either measure would result in a similar prevalence of diabetes in the population.⁴

In 2009, the International Expert Committee Report on the Role of the A1C Assay in the Diagnosis of Diabetes concluded that data from diverse populations provided strong justification for assigning a hemoglobin A1c (HbA1c) cut point of ≥6.5% for the diagnosis of diabetes based on an increase in the prevalence of moderate retinopathy.⁵ The HbA1c assay is at least as good at defining the level of hyperglycemia at which retinopathy prevalence increases, has appreciably superior technical attributes (including less pre-analytic instability and less biologic variability), and is more clinically convenient than FPG and the OGTT.⁵

In the past several years, however, many studies have reported on some inherent limitations of the HbA1c test. Both intra-individual variations, such as erythrocytic process, and inter-individual variations, such as race and age, are known to influence HbA1c.²

Ethnic disparities between HbA1c and mean glucose levels have been shown in several studies. In the Diabetes Prevention Program, which studied over 3,000 participants with impaired glucose tolerance, it was found that for the same mean glucose levels, the HbA1c level was higher among Hispanics, Asians, American Indians, and blacks than in whites.⁶ However, it is important to note that these findings do not support increasing the diagnostic threshold of HbA1c in black persons because data from the National Health and Nutrition Examination Survey (NHANES) (2005–2008) database showed that the prevalence of retinopathy begins to increase at a lower HbA1c level in black Americans than in white Americans.⁷

As regards to the effects of age on HbA1c, there is growing evidence to show that HbA1c levels increase with increasing age.^8,9 However, higher HbA1c levels with increasing age cannot be attributed to unrecognized diabetes or prediabetes, postprandial hyperglycemia, or insulin resistance. This was demonstrated by Dubowitz et al.,¹⁰ who performed a cross-sectional analysis of data from adults without known diabetes in the Screening for Impaired Glucose Tolerance study (2005–2008) (n = 1,573) and NHANES (2005–2006) (n = 1,184). They found that in multivariate analyses of subjects with normal glucose tolerance, the relationship between age and HbA1c remained significant (P < 0.001) after adjustment for covariates including race, body mass index, waist circumference, sagittal abdominal diameter, triglyceride/high-density lipoprotein ratio, and fasting and 2-h PG and other glucose levels (as assessed by an OGTT). In this analysis, 10 additional years of age was associated with a rise in HbA1c levels of approximately 0.07%, and the authors stated that using this calculation, the HbA1c level of an 80-year-old individual with normal glucose tolerance would be 0.35% greater than that of a 30-year-old individual with normal glucose tolerance with the same glucose levels. Furthermore, in this analysis, the specificity of HbA1c-based diagnostic criteria was also found to decrease substantially with increasing age. These findings have potential implications for the use of HbA1c to guide diagnosis and management in older individuals.¹¹

In this issue of Diabetes Technology & Therapeutics Zou et al.¹² expand on the findings of Dubowitz et al.¹⁰ by demonstrating an age-dependent decline in the diagnostic accuracy of HbA1c for diabetes screening in a Chinese population. In addition, they show that the magnitudes of differences in levels of FPG, 2-h PG, glycated albumin, and HbA1c between those with and without diabetes all decrease with increasing age (the glycemic gap). They hypothesize that the age-dependent decline in the diagnostic accuracy of HbA1c for diabetes screening is attributable (at least partly) to a reduced glycemic gap between those with and without diabetes in older individuals (age group of 55–75 years).

The population studied was from a suburban district of Beijing (Pinggu), China. Individuals without known diabetes were screened from March 2012 to May 2013, and after people with anemia and chronic kidney disease were excluded, 3,050 individuals were included in the final analysis. OGTTs were conducted, and HbA1c was measured in all participants. They were then divided in to three groups by age: the first tertile (25–41 years), the second tertile (42–54 years), and the third tertile (55–75 years). The authors calculated the area under the curve (AUC) of the receiver operating characteristic curve (ROC) for HbA1c, FPG, and 2-h PG to detect diabetes in each group. Compared with young participants, the diagnostic accuracy for detecting diabetes with HbA1c was reduced in middle-aged and old people (the median [interquartile range] ROC AUC was 0.958 [0.915, 1.000], 0.891 [0.852, 0.930], and 0.861 [0.821, 0.901] in young, middle-aged and older people, respectively; P = 0.005).

To test their hypothesis that the decreased accuracy of HbA1c for detecting OGTT-defined diabetes in older individuals is associated with a narrowed difference between glucose levels of individuals with and without diabetes in older age groups (reduced glycemic gap), they mathematically modeled the data. First, they calculated the mean PG level of each individual (mean of FPG and 2-h PG) and divided individuals into the nondiabetes group and the diabetes group by tertiles according to the mean PG level. Then they artificially created a “large difference group”: those from the highest tertile in the diabetes group and the lowest tertile from from the nondiabetes group. The “small difference group” comprised those from the highest-tertile nondiabetes group and the lowest-tertile diabetes group. Finally, they divided the “large difference group” and the “small difference group” into young, middle-aged, and older-aged groups.

As expected, the difference in glycemia (change in FPG and change in 2-h PG) in the “large difference group” was significantly higher than that in the “small difference group.” Furthermore, in the “large difference group,” the ROC AUC for detecting glucose-defined diabetes was 1.000 [1.000, 1.000], 1.000 [1.000, 1.000], and 0.999 [0.998, 1.000] in young, middle-aged, and older participants, respectively (P = 0.408). On the other hand, in the “small difference group,” the ROC AUC was 0.829 [0.670, 0.988], 0.800 [0.668, 0.8318], and 0.685 [0.600, 0.770] in the three consecutive age groups (P = 0.262). Thus, the use of mathematical modeling to artificially increase the change in FPG and 2-h PG differences improves the diagnostic accuracy of the HbA1c level in all age groups, including older individuals, whereas an artificial decrease of the differences in change in FPG and 2-h PG decreases the diagnostic accuracy in all age groups, including the younger individuals.

There is a growing epidemic of diabetes across the world and a pressing need to identify and treat patients with diabetes early in order to reduce the burden of diabetes complications. Diagnosing and treating diabetes, especially in the elderly population, must be approached with caution given the characteristics of HbA1c to increase with advancing age independent of glycemia, and, on the other hand, its declining sensitivity to diagnose diabetes in older individuals, perhaps because of a decreased glycemic gap, as suggested in the study by Zou et al.¹²

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Ahmed

: History of diabetes mellitus. Saudi Med J, 2002; 23:373–378.

Sacks

: A1C versus glucose testing: a comparison. Diabetes Care, 2001; 34:518–523.

National Diabetes Data Group: Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes, 1979; 28:1039–1057.

The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus: Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care, 1997; 20:1183–1197.

International Expert Committee: International Expert Committee report on the role of the A1C assay in the diagnosis of diabetes. Diabetes Care, 2009; 32:1327–1334.

Herman

, Ma

, Uwaifo

, et al.: Differences in A1C by race and ethnicity among patients with impaired glucose tolerance in the Diabetes Prevention Program. Diabetes Care, 2007; 30:2453–2457.

Tsugawa

, Mukamal

, Davis

, et al.: Should the hemoglobin A1c diagnostic cutoff differ between blacks and whites? A cross-sectional study. Ann Intern Med, 2012; 157:153–159.

Davidson

, Schriger

: Effect of age and race/ethnicity on HbA1c levels in people without known diabetes mellitus: implications for the diagnosis of diabetes. Diabetes Res Clin Pract, 2010; 87:415–421.

Pani

, Korenda

, Meigs

, et al.: Effect of aging on A1C levels in individuals without diabetes: evidence from the Framingham Offspring study and the National Health and Nutrition Examination Survey 2001–2004. Diabetes Care, 2008; 31:1991–1996.

10.

Dubowitz

, Xue

, Long

, et al.: Aging is associated with increased HbA1c levels, independently of glucose levels and insulin resistance, and also with decreased HbA1c diagnostic specificity. Diabet Med, 2014; 31:927–935.

11.

Selvin

, Coresh

, Brancati

: The burden and treatment of diabetes in elderly individuals in the U.S. Diabetes Care, 2006; 29:2415–2419.

12.

Zou

, Li

, Cai

, et al.: Decreased glycemic difference between diabetes and nondiabetes in the elderly leads to the reduced diagnostic accuracy of hemoglobin A1c for diabetes screening in an aged Chinese population. Diabetes Technol Ther, 2016; 18:226–232.