Insufficient sensitivity of hemoglobin A1C determination in diagnosis or screening of early diabetic states

doi:10.1016/j.metabol.2010.06.017

Metabolism

Volume 60, Issue 1, January 2011, Pages 86-91

https://doi.org/10.1016/j.metabol.2010.06.017 Get rights and content

Abstract

An International Expert Committee made recommendations for using the hemoglobin A_1C (A1C) assay as the preferred method for the diagnosis of diabetes in nonpregnant individuals. A concentration of at least 6.5% was considered as diagnostic. It is the aim of this study to compare the sensitivity of A1C with that of plasma glucose concentrations in subjects with early diabetes or impaired glucose tolerance (IGT). We chose 2 groups of subjects who had A1C not exceeding 6.4%. The first group of 89 subjects had family histories of diabetes (MODY or type 2 diabetes mellitus) and had oral glucose tolerance test (OGTT) and A1C determinations. They included 36 subjects with diabetes or IGT and 53 with normal OGTT. The second group of 58 subjects was screened for diabetes in our Diabetes Clinic by fasting plasma glucose, 2-hour plasma glucose, or OGTT and A1C; and similar comparisons were made. Subjects with diabetes or IGT, including those with fasting hyperglycemia, had A1C ranging from 5.0% to 6.4% (mean, 5.8%). The subjects with normal OGTT had A1C of 4.2% to 6.3% (mean, 5.4%), or 5.5% for the 2 groups. The A1C may be in the normal range in subjects with diabetes or IGT, including those with fasting hyperglycemia. Approximately one third of subjects with early diabetes and IGT have A1C less than 5.7%, the cut point that the American Diabetes Association recommends as indicating the onset of risk of developing diabetes in the future. The results of our study are similar to those obtained by a large Dutch epidemiologic study. If our aim is to recognize early diabetic states to apply effective prophylactic procedures to prevent or delay progression to more severe diabetes, A1C is not sufficiently sensitive or reliable for diagnosis of diabetes or IGT. A combination of A1C and plasma glucose determinations, where necessary, is recommended for diagnosis or screening of diabetes or IGT.

Section snippets

Research design and methods

We include 2 separate groups of subjects in this report. One group is composed of subjects from a large, long-term study of the natural history of diabetes. In this study, we have performed routine OGTTs in apparently healthy first-degree relatives of known diabetic patients. We selected 89 subjects who had an OGTT and an A1C determination on the same day and who had an A1C concentration not exceeding 6.4%. As the cut point for a diagnosis of diabetes has been proposed as greater than or equal

Biochemical assays

All plasma samples were collected on ice; spun; and, if not processed immediately, stored at −70°C until assayed. Plasma glucose was measured on a Cobas Mira Plus Analyzer (Roche Diagnostics, Indianapolis, IN) using a hexokinase method. Interassay variation is 3.15% at 84 mg/dL and 2.8% at 292 mg/dL (n = 224). Hemoglobin A_1C was measured in whole blood immediately using the assay kit Unimate 3 from Roche Diagnostics. Hemoglobin A_1C and total hemoglobin are determined from hemolysate, prepared

Results

As shown in Table 3, the 36 subjects in group 1 with either diabetes (n = 28) or IGT (n = 8) had a mean A1C concentration of 5.8%. This was true for the 11 subjects with fasting hyperglycemia (group A: FPG ≥126 mg/dL; mean, 142 mg/dL) as well as for the 17 subjects with FPG less than 126 mg/dL (groups B and C). The A1C concentrations were as low as 5.0% in some subjects. In group A, 8 of 11 subjects had A1C concentrations less than 6.0%. Similar results were found in the subjects with IGT

Discussion

This study addresses the detection of very early defects in glucose metabolism in subjects with very mild diabetes and IGT who are at high risk of developing more severe diabetes with complications but had A1c concentration not exceeding 6.4%. Although the number of subjects studied is limited, the data and conclusions derived from them may have an important impact on the recognition of diabetes in its early stages. A larger number of subjects with similar mild abnormalities will have to be

Acknowledgment

This work used the Biochemistry Core of the Michigan Diabetes Research and Training Center funded by DK020572 from the National Institute of Diabetes and Digestive and Kidney Diseases; US Public Health Service Grant M-01-RR-00042 to the General Clinical Research Center; and CTSA Grant UL1RR024986, all at the University of Michigan. We thank Yugandhar Kalagara for his help in preparing the tables.

References (23)

InoueK. et al.
The combination of fasting plasma glucose and glycosylated hemoglobin predicts type 2 diabetes in Japanese workers
Diabetes Res Clin Pract
(2007)
International Expert Committee
International Expert Committee report on the role of the A1C assay in the diagnosis of diabetes
Diabetes Care
(2009)
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)
World Health Organization
Definition, diagnosis and classification of diabetes mellitus and its complications: report of a WHO consultation. Part 1: diagnosis and classification of diabetes mellitus
(1999)
HermanW.H.
Diabetes epidemiology: guiding clinical and public health practice: the Kelly West Award lecture
Diabetes Care
(2006)
MohanV. et al.
A1C cut points to define various glucose intolerance groups in Asian Indians
Diabetes Care
(2010)
CowieC.C. et al.
Prevalence of diabetes and high risk for diabetes using hemoglobin A1c criteria in the U.S. population in 1988-2006
Diabetes Care
(2010)
ChristensenD.L. et al.
Moving to an A1C-based diagnosis of diabetes has a different impact on prevalence in different ethnic groups
Diabetes Care
(2010)
EngelgauM.M. et al.
Comparison of fasting and 2-hour glucose and HbA1c levels for diagnosing diabetes. Diagnostic criteria and performance revisited
Diabetes Care
(1997)
SaudekC.D. et al.
A new look at screening and diagnosing diabetes mellitus
J Clin Endocrinol Metab
(2008)

MosesR.G. et al.

Point/counterpoint: point: universal screening for gestational diabetes mellitus

Diabetes Care

(2009)

Cited by (34)

Use of the new guidelines on an earlier age threshold of 35 years for diabetes screening can identify an additional 6.3 million undiagnosed individuals with diabetes and 72.3 million individuals with prediabetes among Chinese adults: An analysis of a nationally representative survey
2022, Metabolism: Clinical and Experimental
Citation Excerpt :
The three glycemic measurements might identify different numbers of additional individuals with diabetes or prediabetes. It was noteworthy that although OGTT was performed as the initial screening test, it could detect additional cases not captured by other metrics [29]. For instance, it uncovered 5.6 % and 20.7 % of diabetes and prediabetes not meeting diagnostic criteria based on HbA1c in a previous study [30].
Young-onset diabetes has been increasingly prevalent in China and most of the young patients with diabetes remain undiagnosed. Recently, the American Diabetes Association (ADA) updated their screening criteria and turned down the age threshold of diabetes screening from 45 years to 35 years, which highlighted the importance of identifying young individuals with diabetes. Herein, we aimed to evaluate the clinical relevance of updated ADA screening recommendations in Chinese adults and the metabolic features and risk factor profiles of these newly diagnosed individuals.
Using a complex, multistage, probability sampling design, we analyzed data from a nationally representative sample of 98,658 Chinese adults in 2010. Participants without previously diagnosed diabetes were included into the present study. We calculated the proportion of individuals with diabetes eligible for screening and the number needed to screen (NNS) to identify one patient with diabetes by age groups.
Setting an earlier age threshold of diabetes screening can identify additional 6.3 million patients with diabetes and 72.3 million individuals with prediabetes, and the proportion of identified individuals increased more in rural, underdeveloped, and central areas. The NNS in Chinese adults dropped significantly from 28 in 30–34 age group to 15 in 35–45 years of age and remained low afterwards. The undiagnosed patients with diabetes who met the new screening age threshold of ADA recommendation were characterized by younger age, lower blood pressure and blood lipids, but higher proportion of overweight and higher level of insulin resistance, and tended to have an unhealthy diet habit, including low intake of fruits and vegetables and high intake of sugar-sweetened beverages, compared to those aged over 45 years.
The new age threshold of 35 years for diabetes screening would reduce the proportion of undiagnosed diabetes with high cost-effectiveness, given the NNS for a positive test result was much lower in 35–45 age group comparing to the lower age group in Chinese adults.
Plasma glycemic measures and fecundability in a Singapore preconception cohort study
2021, Fertility and Sterility
To examine the association between plasma glycemia in women attempting to conceive and fecundability, as measured by time to pregnancy.
Prospective preconception population-based study.
Hospital.
Asian preconception women, 18–45 years old, attempting conception for ≤12 cycles at study entry.
None.
We ascertained time to pregnancy within a year of glycemic assessment in menstrual cycles. We estimated fecundability ratios (FRs) and 95% confidence intervals using discrete-time proportional hazards models, adjusting for age, ethnicity, education, body mass index, and cycle regularity and accounting for left truncation and right censoring.
We studied a population sample of 766 women from the Singapore Preconception Study of Long-Term Maternal and Child Outcomes prospective cohort. Compared with women with normoglycemia, women with dysglycemia (prediabetes and diabetes, defined by the American Diabetes Association) had a lower FR (0.56). Compared with the respective lowest quintiles, women in the highest quintile of fasting glucose (≥5.1 mmol/L) had an FR of 0.60, while women in the highest 2-hour postload glucose quintile (≥6.9 mmol/L) had an FR of 0.66. Overall, the FRs decreased generally across the range of fasting and 2-hour plasma glucose. Glycated hemoglobin was not associated with fecundability.
Increasing preconception plasma glucose is associated with reduced fecundability, even within the normal range of glucose concentrations.
NCT03531658.
Determinaciones plasmáticas de glucemia y fecundidad en un estudio preconcepcional de cohortes en Singapur.
Estudiar la asociación entre la glucemia plasmática en mujeres que buscan gestación y la fecundidad medida por tiempo hasta conseguir el embarazo.
Estudio poblacional prospectivo preconcepcional.
Hospitalario.
Mujeres asiáticas antes de la concepción, entre 18 y 45 años, que intentan concebir durante ≤12 ciclos al ingresar al estudio.
Ninguna.
Determinamos el tiempo hasta la gestación en el plazo de un año de mediciones glucémicas en los ciclos menstruales. Estimamos las tasas de fecundidad (FRs) y los intervalos de confianza del 95% mediante modelos de riesgo proporcional de tiempo discreto, ajustando por edad, etnia, nivel educativo, índice de masa corporal y regularidad del ciclo menstrual, y teniendo en cuenta el truncamiento izquierdo y la censura derecha.
Estudiamos una muestra poblacional de 766 mujeres procedentes de la cohorte prospectiva del estudio preconceptivo de Singapur sobre los resultados maternos y de hijos a largo plazo (“Singapore Preconception Study of Long-Term Maternal and Child Outcomes”). Las mujeres con disglucemia (prediabetes y diabetes definidas por la asociación americana de diabetes) tuvieron una menor FR (0.56), comparadas con las mujeres normoglucémicas. Las mujeres en el quintil con la mayor glucemia en ayunas (≥5.1 mmol/L) tuvieron una FR de 0.60, mientras que las mujeres en el quintil con mayor glucemia a las 2 horas tras la sobrecarga (≥6.9 mmol/L) tuvieron una FR de 0.66, comparadas con los respectivos quintiles de menor glucemia. En general, las FRs disminuyeron generalmente a lo largo del rango de la glucemia plasmática en ayunas y a las 2 horas. La hemoglobina glicosilada no estuvo asociada con la fecundidad.
El aumento de los niveles glucémicos plasmáticos está asociado con la reducción de la fecundidad, incluso dentro de los rangos normales de concentración de glucosa plasmática.
Optimal hemoglobin A1C cutoff value for diabetes mellitus and pre-diabetes in Pudong New Area, Shanghai, China
2018, Primary Care Diabetes
Due to the diversity of the Chinese population, it requires considerable research to evaluate HbA1c diagnostic threshold for diagnosis of hyperglycemia.
We included 7909 subjects aged ≥15 without known diabetes from the baseline of Pudong community cohort in 2013. Participants took oral glucose tolerance test (OGTT) and HbA1c assay. Receiver operating characteristic curve determined the HbA1c threshold in the diagnosis of hyperglycemia.
The optimal HbA1C threshold for diagnosing newly diagnosed diabetes (NDD) and pre-diabetes in this population was 6.0% (AUC = 0.798, 95%CI: 0.779–0.818) and 5.6% (AUC = 0.655, 95%CI: 0.638–0.671). When compared with elderly age group (≥70 years), HbA1c for detecting NDD performed better in youth (15–39 years: P = 0.003, 40–49 years: P < 0.001). There were 13.81% and 13.34% of participants would be newly detected as NDD and pre-diabetes via HbA1c criteria; meanwhile 3.20% and 15.52% diagnosed as NDD and pre-diabetes by OGTT criteria would be missed diagnosis.
The optimal HbA1c thresholds for NDD and pre-diabetes were lower than ADA criteria. It is necessary to carefully consider whether choose HbA1c as a diagnostic criterion or combine two diagnostic standards. Age-specific diagnostic thresholds should be considered when HbA1c was recommended as diagnostic standard.
Detecting prediabetes among Hispanics/Latinos from diverse heritage groups: Does the test matter? Findings from the Hispanic Community Health Study/Study of Latinos
2017, Preventive Medicine
Citation Excerpt :
IFG reflects increased hepatic glucose output, which leads to fasting hyperglycemia (Inzucchi, 2012), and is assessed by fasting plasma glucose (FPG). Hemoglobin A1c (HbA1c) correlates directly with the preceding 2–3 month-mean plasma glucose levels, and has been shown to be elevated during states of intermediate glucose homeostasis (Inzucchi, 2012; Sacks, 2011; Olson et al., 2010; Cohen et al., 2010; Lorenzo et al., 2010; Selvin et al., 2011; Fajans et al., 2011). In 2010, the American Diabetes Association (ADA) recommended cut points for the diagnosis of diabetes mellitus and prediabetes based on FPG, OGTT, and HbA1c (American Diabetes Association, 2010).
The objectives of this analysis were to compare the ability of fasting plasma glucose (FPG), post oral load plasma glucose (2hPG), and hemoglobin A_1c (HbA_1c) to identify U.S. Hispanic/Latino individuals with prediabetes, and to assess its cardiovascular risk factor correlates.
This is a cross-sectional analysis of baseline data from 15,507 adults without self-reported diabetes mellitus from six Hispanic/Latino heritage groups, enrolled in the Hispanic Community Health Study/Study of Latinos, which takes place in four U.S. communities. The prevalence of prediabetes was determined according to individual or combinations of ADA-defined cut points: FPG = 5.6–7.0 mmol/L, 2hPG = 7.8–11.1 mmol/L, and HbA_1c = 5.7%–6.4% (39–46 mmol/mol). The sensitivity of these criteria to detect prediabetes was estimated. The prevalence ratios (PRs) for selected cardiovascular risk factors were compared among alternative categories of prediabetes versus normoglycemia [FPG < 5.6 mmol/L and 2hPG < 7.8 mmol/L and HbA_1c < 5.7% (39 mmol/mol)].
Approximately 36% of individuals met any of the ADA prediabetes criteria. Using 2hPG as the gold standard, the sensitivity of FPG was 40.1%, HbA_1c was 45.6%, and that of HbA_1c + FPG was 62.2%. The number of significant PRs for cardiovascular risk factors was higher among individuals with isolated 2hPG = 7.8–11.1 mmol/L, FPG = 5.6–7.0 mmol/L + HbA_1c = 5.7%–6.4%, or those who met the three prediabetes criteria.
Assessing FPG, HbA_1c, and cardiovascular risk factors in Hispanics/Latinos at risk might enhance the early prevention of diabetes mellitus and cardiovascular complications in this young and growing population, independent of their heritage group.
Performance of A1C versus OGTT for the diagnosis of prediabetes in a community-based screening
2016, Endocrine Practice
Objective: Reliable identification of individuals at risk for developing diabetes is critical to instituting preventative strategies. Studies suggest that the accuracy of using hemoglobin A1c as a sole diagnostic criterion for diabetes may be variable across different ethnic groups. We postulate that there will be lack of concordance between A1c and the oral glucose tolerance test (OGTT) for diagnosing prediabetes across Hispanic and non-Hispanic white (NHW) populations.
Methods: A total of 218 asymptomatic adults at risk for type 2 diabetes (T2D) were assessed with A1c and OGTT for the diagnosis of prediabetes. Glucose homeostasis status was assigned as no diabetes (A1c <5.7% [39 mmol/mol]), prediabetes (A1c 5.7 to 6.4% [46 mmol/mol]), and T2D (A1c >6.4% [46 mmol/mol]). Inclusion criteria were age >18 years and at least one of the following: a family history of diabetes, a history of gestational diabetes, Hispanic ethnicity, non-Caucasian race, or obesity. Subjects received a fasting 75-g OGTT and A1c on the same day. Bowker's test of symmetry was employed to determine agreement between the tests.
Results: Data from 99 Hispanic patients and 79 NHW patients were analyzed. There was no concordance between A1c and OGTT for Hispanic (P =.002) or NHW individuals (P =.003) with prediabetes.
Conclusion: A1c is discordant with OGTT among Hispanic and NHW subjects for the diagnosis of prediabetes. Sole use of A1c to designate glycemic status will result in a greater prevalence of prediabetes among Hispanic and NHW New Mexicans.
Abbreviations:
A1c = hemoglobin A1c
BMI = body mass index
CDC = Centers for Disease Control
CI = confidence interval
FPG = fasting plasma glucose
NHW = non-Hispanic white
OGTT = oral glucose tolerance test
T2D = type 2 diabetes
WHO = World Health Organization
Current aspects in hemoglobin A1c detection: A review
2015, Clinica Chimica Acta
Type 2 diabetes mellitus (T2DM) is a pressing health issue that threatens global health and the productivity of populations worldwide. Despite its long-recognized role in diabetes management, glycated hemoglobin (HbA1c) only received WHO endorsement as a T2DM diagnostic tool in 2011. Although conventional plasma-specific tests have long been utilized to diagnose T2DM, the public should be informed that plasma-specific tests are not markedly better than HbA1c tests, particularly in terms of variability and convenience for diagnosing diabetes. In the midst of the debates associated with establishing HbA1c as the preeminent diabetes diagnostic tool, unceasing efforts to standardize HbA1c tests have played an integral part in achieving more efficient communication from laboratory to clinical practice and thus better diabetes care. This review discusses the current status of HbA1c tests in the diagnosis, prevention, treatment and management of T2DM across the globe, focusing on increasing the recognition of glycated hemoglobin variants with effective utilization of different HbA1c methods, updating the current status of HbA1c standardization programs, tapping into the potential of POC analyzers to establish a cost-effective HbA1c test for diabetes care, and inspiring the advancement of HbA1c biosensors for future clinical usage.

View all citing articles on Scopus

: A preliminary abstract of this work was published in Diabetes 58:2245, 2009, Supplement 1.

View full text

Published by Elsevier Inc.

Insufficient sensitivity of hemoglobin A1C determination in diagnosis or screening of early diabetic states

Abstract

Section snippets

Research design and methods

Biochemical assays

Results

Discussion

Acknowledgment

Diabetes Res Clin Pract

International Expert Committee report on the role of the A1C assay in the diagnosis of diabetes

Diabetes Care

Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus

Diabetes Care

Definition, diagnosis and classification of diabetes mellitus and its complications: report of a WHO consultation. Part 1: diagnosis and classification of diabetes mellitus

Diabetes epidemiology: guiding clinical and public health practice: the Kelly West Award lecture

Diabetes Care

A1C cut points to define various glucose intolerance groups in Asian Indians

Diabetes Care

Prevalence of diabetes and high risk for diabetes using hemoglobin A1c criteria in the U.S. population in 1988-2006

Diabetes Care

Moving to an A1C-based diagnosis of diabetes has a different impact on prevalence in different ethnic groups

Diabetes Care

Comparison of fasting and 2-hour glucose and HbA1c levels for diagnosing diabetes. Diagnostic criteria and performance revisited

Diabetes Care

A new look at screening and diagnosing diabetes mellitus

J Clin Endocrinol Metab

Point/counterpoint: point: universal screening for gestational diabetes mellitus

Diabetes Care

Insufficient sensitivity of hemoglobin A_1C determination in diagnosis or screening of early diabetic states