A Single A1C ≥ 6.5% Accurately Identifies Type 2 Diabetes/Impaired Glucose Tolerance in African Americans

Abstract

Objective: In 2010, the American Diabetes Association revised its criteria for the diagnosis of diabetes to include A1C ≥ 6.5%; however, this has remained controversial, particularly for African Americans. The objective of this pilot study was to examine the usefulness of a single A1C determination in comparison with a same day 2-hour oral glucose tolerance test to diagnose type 2 diabetes in African Americans. Methods: In sum, 195 oral glucose tolerance tests and A1Cs were obtained on the same day from 77 overweight and obese African American women and 6 men over a period of 15 months. Results: A1C ≥ 6.5% was present in 31 of 195 patients, with 15 of these having type 2 diabetes by oral glucose tolerance test, another 12 having impaired glucose tolerance, and 4 having normal glucose tolerance. This gives a sensitivity of 50% and a specificity of 90%, with a positive predictive value of 48% and a negative predictive value of 91%. A1C ≤ 5.6%, proposed by the American Diabetes Association to indicate normal glucose tolerance, was present in only 28 patients, 10 (35.7%) of whom had normal glucose tolerance, whereas 18 (64.3%) had either impaired glucose tolerance (15 patients) or type 2 diabetes (3 patients). Fasting plasma glucose ≥ 126 mg/dL was present in 5 of 29 patients with type 2 diabetes (sensitivity, 17.2%; specificity, 100%). Conclusions: First, A1C ≥ 6.5% was a good “rule in” value to identify impaired glucose tolerance and type 2 diabetes (ie, patients at high risk for micro- and macrovascular complications). Second, A1C ≤ 5.6% did not rule out impaired glucose tolerance or type 2 diabetes. Last, fasting plasma glucose ≥ 126 mg/dL detected less than 1 in 5 cases with type 2 diabetes.

Keywords

A1C oral glucose tolerance test fasting plasma glucose African Americans

Several years usually pass between onset and diagnosis of type 2 diabetes. During this time, hyperglycemia and other metabolic abnormalities remain untreated and promote the development of micro- and macrovascular complications.^1,2 To facilitate earlier recognition of diabetes and to delay or prevent complications, in 2010 the American Diabetes Association revised its criteria for the diagnosis of diabetes to include A1C ≥ 6.5%, following the recommendations of an International Expert Committee.^3,4 Other criteria accepted by the association for the diagnosis of diabetes include fasting plasma glucose (FPG) ≥ 126 mg/dL and 2-hour plasma glucose ≥ 200 mg/dL during an oral glucose tolerance test (OGTT). Advantages of the A1C compared to the FPG or 2-hour plasma glucose are (1) convenience, as individuals are not required to fast, (2) less day-to-day variability and more preanalytic stability, and (3) results better reflecting longer-term glycemic control.⁴

However, several recent studies^5-7 have raised doubts about the validity of A1C as a diagnostic test. These studies suggested that A1C ≥ 6.5% will miss many cases of OGTT-confirmed type 2 diabetes. Moreover, there is little information on the ability of A1C to diagnose type 2 diabetes in African Americans. This is particularly important because higher A1C levels have been reported in African Americans than in Caucasians.^8-10 For instance, in the Diabetes and Prevention Program,⁸ the mean A1C after adjusting for independent predictors was 6.19% ± 0.59% in black versus 5.80% ± 0.44% in white participants. This difference was independent of glucose concentration; it persisted after adjustment for other factors likely to affect glycemia; and it has been suggested to be due to differences in hemoglobin glycation or red cell survival.^11,12

Therefore, it was the objective of this pilot study to examine the usefulness of a single A1C determination in comparison to a same-day 2-hour OGTT to diagnose type 2 diabetes in a group of predominantly female and overweight/obese African American participants.

Methods

We performed 195 same-day OGTT and A1C studies on 77 African American women and 6 men. The clinical characteristics of the study participants are shown in Table 1. All were overweight or obese without known type 2 diabetes or anemia, and all were enrolled in a clinical trial to examine the impact of a telemedicine program on weight maintenance (NCT 00373230; clinical trials.gov). All participants underwent a 12-week church-based behavioral weight loss program and were followed for 1 year either by attending face-to-face meetings or by utilizing a web-based telemedicine program. At baseline, all participants had a 2-hour OGTT (75 g of glucose) and an A1C test performed on the same day. OGTTs and the same-day A1Cs were repeated in 61 participants at 12 weeks (following the weight loss program) and in 51 during their final visit (following the 12-month weight maintenance period). Of the 195 OGTTs, 181 were performed in women and 14 in men.

Table 1.

Baseline Characteristics of Study Participants^a

N = 83	77 women, 6 men
Age, y	53 ± 10.4
Weight, lb	210 ± 34.8
Body mass index, kg/m²	35.4 ± 5.0
Family history of type 2 diabetes	40 of 83 (48.2%)
A1C (%)	6.0 ± 0.5
Oral glucose tolerance test, mg/dL
Fasting plasma glucose	101.5 ± 12.4
2-hour plasma glucose	161.3 ± 43.8
Blood pressure, mm/Hg
Systolic	129.1 ± 15.9
Diastolic	78.8 ± 9.4

Mean ± SD

All OGTTs were performed in the morning after an overnight fast. Blood was collected from an indwelling anticubital vein catheter at 0, 30, 60, 90, and 120 minutes. Utilizing current American Diabetes Association criteria,³ normal glucose tolerance was defined as FPG < 100 mg/dL and a 2-hour plasma glucose < 140 mg/dL. Impaired glucose tolerance (IGT) was defined as FPG ≥ 100 mg/dL but < 126 mg/dL and/or a 2-hour plasma glucose ≥ 140 mg/dL but < 200 mg/dL. Type 2 diabetes was diagnosed when FPG ≥ 126 mg/dL and/or 2-hour plasma glucose ≥ 200 mg/dL.

All participants provided written informed consent. The study protocol was approved by the Temple University Institutional Review Board.

A1C was measured using a Centers for Disease Control and Prevention–approved automated point-of-care analyzer (DCA 2000, Bayer Corporation, Tarrytown, New York). This technique uses a monoclonal antibody, which recognizes the glycated N-terminus of the β chain of hemoglobin. Correlation coefficients ranged from 0.95 to 0.99 comparing DCA 2000 with other approved methods, including the DCCT/EDIC central laboratory at the University of Minnesota.¹³ The recommended reference range was 4.3% to 5.7%.¹⁴

Plasma glucose was measured by a glucose analyzer based on the glucose oxidase method, and serum insulin, by radioimmunoassay based on an antiserum with minimal (< 0.2%) cross-reactivity with proinsulin (Linco Research, St Charles, Missouri).

Results

A1C ≥ 6.5% indicates type 2 diabetes or IGT in African Americans

In sum, 195 studies were done over a period of 15 months, during which some participants underwent weight changes while others were weight stable (ie, conditions commonly seen in real life). Of the 31 patients with A1C ≥ 6.5%, 15 had type 2 diabetes by OGTT; 12 had IGT; and 4 had normal glucose tolerance (sensitivity, 50%; specificity, 90%; positive predictive value, 48%; and negative predictive value, 91%). For those individuals at risk of complications of type 2 diabetes (IGT + type 2 diabetes), A1C ≥ 6.5% had a positive predictive value of 89%.

A1C ≤ 5.6% does not exclude type 2 diabetes or IGT

Twenty-eight patients had A1C ≤ 5.6%, the American Diabetes Association recommended cut point for normal risk to develop type 2 diabetes. Of these 28, only 10 had normal OGTTs; 15 had IGT; and 3 had type 2 diabetes. Therefore, using an A1C cut point of ≤ 5.6% would have misdiagnosed 64.7% as having normal glucose tolerance when in fact they had either IGT or type 2 diabetes (Table 2).

Table 2.

A1C vs Oral Glucose Tolerance Test in 195 Studies Performed in Overweight/Obese African Americans

	Oral Glucose Tolerance Test
A1C, %	Normal Glucose Tolerance^a	Impaired Glucose Tolerance^b	Type 2 Diabetes	Total
≤ 5.6	10	15	3	28
5.7%-6.4	27	97	12	136
≥ 6.5	4	12	15	31
Total	41	124	30

Fasting plasma glucose < 100 mg/dL; 2-hour plasma glucose < 140 mg/dL.

Fasting plasma glucose, 100-125 mg/dL; 2-hour plasma glucose, 140-199 mg/dL.

Fasting plasma glucose ≥ 126 mg/dL; 2-hour plasma glucose ≥ 200 mg/dL.

Comparison of FPG vs 2-hour plasma glucose

Of the 3 tests available to screen for diabetes (FPG, 2-hour plasma glucose, A1C), the FPG is the most widely used. Our data allowed us to compare FPG with 2-hour plasma glucose with respect to their respective sensitivities and specificities to detect type 2 diabetes. FPG ≥ 126 mg/dL detected type 2 diabetes in 5 of the 29 patients with diabetic OGTTs (sensitivity, 17.2%; specificity, 100%). FPG < 100 mg/dL was present in 61 patients who had IGT and 8 who had type 2 diabetes (Table 3).

Table 3.

Two-Hour Plasma Glucose vs Fasting Plasma Glucose in 195 Studies Performed in Overweight/Obese African Americans

2-Hr Plasma Glucose, mg/dL	Normal Glucose Tolerance^a	Impaired Glucose Tolerance^b	Type 2 Diabetes^c	Total
< 140	39	31	0	70
140-199	61	35	0	96
≥ 200	8	16	5	29
Total	108	82	5

Fasting plasma glucose < 100 mg/dL.

Fasting plasma glucose, 100 to 125 mg/dL.

Fasting plasma glucose ≥ 126 mg/dL.

Discussion

Early diagnosis of diabetes is key to prevent and/or delay type 2 diabetes and its complications. Currently, the American Diabetes Association recommends 3 tests to screen for and diagnose diabetes: the 2-hour OGTT, the FPG, and most recently, the A1C. The OGTT has long been the gold standard for the diagnosis of diabetes mellitus, but it is expensive, time-consuming, and not popular with physicians and patients. The FPG is rather insensitive and the A1C, controversial,^5-7 particularly in African Americans.^8-10 Here, we have directly compared the 3 tests in African American, mostly female participants at high risk for type 2 diabetes.

We found that the A1C cut point of ≥ 6.5% had a low sensitivity (50%) but high specificity (90%) to detect type 2 diabetes diagnosed by OGTT. Similar sensitivities and specificities have been reported in other studies and populations.^5-10 A large part of this low sensitivity is due to the inability of the A1C to distinguish between type 2 diabetes and IGT. This is not surprising, as the 2 tests measure different factors. A1C determines glycation of hemoglobin, which depends primarily on long-term exposure of red blood cells to glucose. In contrast, the OGTT measures acute blood sugar responses to a glucose load. Thus, the A1C is a “lagging indicator” and can be expected to underdiagnose patients with recent deterioration of glucose tolerance.

Another less-well-recognized reason for the inability of A1C to differentiate between IGT and type 2 diabetes is the current American Diabetes Association recommendation to diagnose diabetes using only the FPG and 2-hour plasma glucose of an OGTT. For instance, in our study, 4 of the 12 patients who had A1C ≥ 6.5% had 2hPG < 200 mg/dl but > 140 mg/dl and thus were classified as having IGT by OGTT but T2DM by A1C criteria. All 4, however, had blood glucoses of > 200 mg/dl at 60 and/or 90 min of the OGTT, i.e. would have been diagnosed as T2DM had the 60 and 90 min values been considered. Regardless, however, of whether these patients were classified as T2DM or IGT, an A1C of ≥ 6.5% seemed to be a reasonable cut point as it identified individuals at high risk for microvascular and macrovascular complications and in need of medical attention while producing only 1 possible misdiagnosis. This was a man who had an A1C value of 6.5% and a normal OGTT result. The reason for this discrepancy is not known but could include recent improvement in glucose tolerance or interpersonal or perhaps even sex differences in hemoglobin glycation.¹¹

A1C ≤ 5.6% has been suggested to indicate normal glucose tolerance.^3,15 However, at least two-thirds of our studies with A1C ≤ 5.6% had either IGT or type 2 diabetes by OGTT. Unfortunately, the relatively low number of studies with normal glucose-tolerant patients did not allow us to establish a reliable A1C cut point to rule out IGT or type 2 diabetes. Thus, used alone, an A1C value of ≤ 5.6% could not be used to rule out IGT and type 2 diabetes. Our results also suggest than an OGTT would be needed to correctly classify glucose tolerance in a large number of individuals with A1C ≤ 5.6%.

FPG ≥ 126 mg/dL is presently the most widely used test to diagnose diabetes. In our study, FPG identified 5 of 29 patients with type 2 diabetes (sensitivity, 17%). The low sensitivity has been well recognized and is not surprising, as type 2 diabetes usually starts postprandially when the insulin need is greatest.¹⁶,¹⁷

A major strength of the study was the fact that we were able to repeat the same tests in the same individuals under different circumstances and at different points in time. Among the individuals with repeat tests, we found that the results remained consistent over time. Limitations of our study include the fact that our sample size was small and contained few patients with normal glucose tolerance. In addition, all were overweight or obese, and men were not well represented.

In summary, we found A1C ≥ 6.5% to be a good “rule in” value to identify IGT and type 2 diabetes in overweight/obese African Americans—that is, patients at high risk for micro- and macrovascular complications—and to be superior to the FPG, which has been utilized widely. However, A1C ≤ 5.6% did not rule out IGT or type 2 diabetes.

Footnotes

Acknowledgements

We thank Constance Harris Crews for typing the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This study was supported by grants from the Commonwealth of Pennsylvania (RFA-0407-04), the American Diabetes Association (1-10-CT-06), and the National Institutes of Health (RO1- DK066003) (all to GB).

Bios

Carol J. Homko, RN, PhD, is an associate research professor in the Department of Medicine, Section of Endocrinology, Metabolism, and Diabetes at the Temple University School of Medicine.

Linda C. Zamora, BSN, RN, CDE, CCRP, is a research nurse on the Clinical Research Center at Temple University Hospital.

Margaret M. Kerper, RN, BSN, CCRP, is a research nurse on the Clinical Research Center at Temple University Hospital.

Maria Mozzoli, BS, is a research assistant on the Clinical Research Center at Temple University Hospital.

Karen Kresge, BS, is a research assistant on the Clinical Research Center at Temple University Hospital.

Guenther Boden, MD, is the Laura H. Carnell Professor of Medicine in the Department of Medicine, Section of Endocrinology, Metabolism, and Diabetes at the Temple University School of Medicine.

References

Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329:977-986.

UK Prospective Diabetes Study Group. Intensive blood-glucose control with sulfonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes. Lancet. 1998;352:837-853.

American Diabetes Association. Position statement: standards of medical care in diabetes—2010. Diabetes Care. 2010;33(suppl 1):S11-S61.

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.

Kramer

Araneta

MRG

Barrett-Connor

. A1C and diabetes diagnosis: the Rancho Bernardo study. Diabetes Care. 2010;33:101-103.

van’t Riet

Alssema

Rijkelijkhuizen

Kostense

Nijpels

Dekker

. Relationship between A1C and glucose levels in the general Dutch population: the New Hoorn Study. Diabetes Care. 2010;33:61-66.

Zhou

Pang

Gao

. Performance of an A1C and fasting capillary blood glucose test for screening newly diagnoses diabetes and pre-diabetes defined by an oral glucose tolerance test in Qingdao, China. Diabetes Care. 2010;33:545-550.

Herman

Uwaifo

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

Lipska

DeRekeneire

Van Ness

. Identifying dysglycemic states in older adults: implications of the emerging use of hemoglobin A1C. J Clin Endocrinol Metab. 2010;95:5289-5295.

10.

Ziemer

Kolm

Weintraub

. Glucose-independent, black-white differences in hemoglobin A1C levels. Ann Intern Med. 2010;152:770-777.

11.

Cohen

Franco

Khera

. Red cell life span heterogenecity in hematologically normal people is sufficient to alter HbA1C. Blood. 2008;112:4284-4291.

12.

Davidson

Schriger

. Effect of age and race/ethnicity on HbA1C levels in people without know diabetes mellitus: implications for the diagnosis of diabetes. Diabetes Res Clin Pract. 2010;87:415-421.

13.

Tamborlande

Kollman

Steffes

. Comparison of fingerstick hemoglobin A1C levels assayed by DCA 2000 with the DCCT/EDIC central laboratory assay: results of a Diabetes Research in Children Network (DirectNet) Study. Pediatr Diabetes. 2005;6:13-16.

14.

Hiar CE. Clinical Summary Report: Clinical Performance of DCA 2000+ Hemoglobin A1C System Six-Minute Assay. New York, NY: Bayer Corporation, 1998.

15.

Cohen

Haggerty

Herman

. HbA1C for the diagnosis of diabetes and prediabetes: Is it time for a mid-course correction? J Clin Endocrinol Metab. 2010;95:5203-5206.

16.

Saad

Knowler

Pettitt

Nelson

Mott

Bennett

. The natural history of impaired glucose tolerance in the Pima Indians. N Engl J Med. 1998;319:1500-1506.

17.

Del Prato

. Loss of early insulin secretion leads to postprandial hyperglycemia. Diabetologia. 2003;46:M2-M8.