Can Diabetes Be Diagnosed in a Single Blood Test?

Miriam E. Tucker

June 18, 2018

It may be possible to diagnose diabetes by measuring fasting blood glucose and hemoglobin A1c (HbA1c) using the same blood sample without requiring a patient to come back for a second visit, new data suggest. 

The findings, from the prospective Atherosclerosis Risk in Communities (ARIC) study, were published online June 18 in Annals of Internal Medicine by Elizabeth Selvin, PhD, of Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, and colleagues.    

The American Diabetes Association (ADA) recommends that a fasting blood glucose, HbA1c, or oral glucose tolerance test be used as an initial screen for diabetes (Diabetes Care. 2018;41:S13-S27). If results exceed diagnostic thresholds, they advise a second test — a repeat of the same or a different one — unless the patient has clear signs of hyperglycemia. That second test, ADA says, should be "performed without delay, using a new blood sample for confirmation."

But according to new findings from a 25-year follow-up of 12,199 ARIC participants without diagnosed diabetes at baseline, using the same sample for fasting blood glucose and HbA1c had a high positive predictive value for later diabetes diagnosis, as well as cardiovascular and kidney disease, and mortality.

"The results of our study suggest that the two tests from one blood sample can provide adequate confirmation of diabetes, potentially allowing a major simplification of current clinical practice guidelines," said Selvin in a press release by her institution.

"Doctors are already doing these tests together — if a patient is obese, for example, and has other risk factors for diabetes, the physician is likely to order tests for both glucose and HbA1c from a single blood sample," she noted. "It's just that the guidelines don't clearly let you use the tests from that one blood sample to make the initial diabetes diagnosis."

"I'm hoping that these results will lead to a change in the clinical guidelines when they are revised in early 2019, which could make identifying diabetes a lot more efficient in many cases," Selvin added.

Innovative and Practical? But Results Require Replication

The results also showed relatively low sensitivity for isolated elevations in fasting glucose or HbA1c at a single time point. Therefore, in such situations a second test at a future time would be necessary "to ensure that cases are not missed, which is consistent with current clinical recommendations," the researchers note.

Moreover, they add, "these results will need to be confirmed in other datasets."

In an accompanying editorial, K.M. Venkat Narayan, MD, and Ram Jagannathan, PhD, of Emory University, Atlanta, Georgia, call the study approach "innovative and practical," whereas "using a repeated test on a new sample from a subsequent visit to confirm diabetes can be logistically cumbersome, inconvenient, and expensive, and can delay patient care."

Narayan and Jagannathan add, "Early detection and diagnosis of diabetes are critical for appropriate initiation of interventions to prevent or delay complications. Simplifying methods to identify and confirm diabetes will facilitate this important clinical and public health priority."

However, like the authors, the editorialists caution the approach "needs replication in other populations before becoming accepted clinical practice."

Confirmed Undiagnosed Diabetes Predicts Later Diagnosis

Enrollment in ARIC, a prospective ongoing study of nearly 16,000 white and black adults in four US communities, began during 1987–1989. For this study, Selvin and colleagues used visit 2, in 1990–1992, as baseline because that was the first time HbA1c was measured.

A total of 12,199 participants without baseline diabetes were used for the incident diabetes analysis.

Confirmed undiagnosed diabetes was defined as HbA1c ≥ 6.5% and fasting glucose ≥ 126 mg/dL (≥ 7.0 mmol/L) among individuals without diagnosed diabetes. Unconfirmed undiagnosed diabetes was defined as no diagnosis of diabetes and only one of the two elevated measures.

Incident diagnosed diabetes was defined by participant's reported use of glucose-lowering medication and/or a physician diagnosis during follow-up.

There were 978 individuals with elevated fasting glucose or HbA1c when tested in 1990–1992, including 39% with confirmed undiagnosed diabetes and 61% with unconfirmed undiagnosed diabetes.

Of note, isolated fasting glucose elevation was four times more common than isolated HbA1c elevation (4% vs 1%).   

Selvin and colleagues found that elevated levels of fasting glucose and HbA1c from a single baseline blood sample had moderate sensitivity (54.9%) but high specificity (98.1%) to identify patients who had diabetes diagnosed during 5 years of follow-up, with specificity increasing to 99.6% at 15 years.

Overall, the adjusted cumulative incidences of diabetes among those with the two positive tests at baseline were 42.0% at 5 years, 97.3% at 15 years, and nearly 100% by 20 years.

In contrast, those with only one elevated result had an adjusted cumulative incidence of only 9.9% at 5 years and 71.7% at 15 years.

At 15 years and after full adjustment, odds ratios (ORs) for incident diagnosed diabetes were 4.17, 4.82, and 15.93 for isolated fasting glucose elevation, isolated HbA1c elevation, and elevations in both (confirmed undiagnosed diabetes) compared with no diabetes, respectively (all significant).

The "confirmed undiagnosed diabetes" status also significantly predicted chronic kidney disease (OR, 1.51), cardiovascular disease (OR, 1.52), peripheral arterial disease (OR, 2.48), and all-cause mortality (OR, 1.49) at 15 years.

"Our findings support clinical use of a combination of HbA1c and fasting glucose levels from a single blood sample to identify cases of undiagnosed diabetes in the population," the researchers reiterate.

Caveats: Necessary "Trade-off Between Sensitivity and Specificity"

Narayan and Jagannathan cite a few study limitations, some of which the authors also noted. First, the 5-year intervals between glucose measurements in the study were far longer than what would be expected in clinical practice. Second, the study included only black and white Americans aged 45 to 64 years, so the generalizability to other patient groups is uncertain. 

Most importantly, they say, concordance between elevated HbA1c and fasting glucose levels from the single samples was approximately 40% in this study, compared with approximately 70% for repeated measurements of fasting glucose at different time points, as based on conventional criteria.

Therefore, the editorialists say, "the criteria proposed by Selvin and colleagues may miss a large number of persons with diabetes. This trade-off between sensitivity and specificity is important because the evidence now favors early intervention for persons with undiagnosed diabetes and even prediabetes."

The ARIC study was funded by the National Institute of Diabetes and Digestive and Kidney Diseases and National Heart, Lung, and Blood Institute. The authors and editorialists have reported no relevant financial relationships.

Ann Intern Med. Published online June 18, 2018. Article, Editorial

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: