Could Single-Sample Testing Confirm a Diabetes Diagnosis?

Elizabeth Selvin, MPH, PhD

While normally a diabetes diagnosis requires multiple sample confirmations from blood work of elevated hemoglobin A1C (HbA1C) or glucose, new data has shown that a single-sample confirmatory diagnosis of diabetes may be possible.

Using data from 12,268 participants without diagnosed diabetes in the Atherosclerosis Risk in Communities (ARIC) study, researchers found that using a single blood sample had a 54.9% sensitivity but 98.1% specificity at the 5-year follow-up point, in its ability provide confirmation of a diabetes diagnosis.

“Our results suggest that this approach could streamline the diagnosis of diabetes. Current guidelines state that the same elevated test needs to be repeated in a new blood sample at a second time point,” Elizabeth Selvin, MPH, PhD, the co-director of the Cardiovascular Disease Epidemiology Training Program at the John’s Hopkins Bloomberg School of Public Health told MD Mag. “This approach is associated with high costs, high patient burden, and can result in delays in treatment. By allowing physicians to make the diagnosis in a single blood sample, at a single point in time, this can make the diagnostic process more efficient.”

“If only 1 test is elevated—a non-confirmed case—then a follow-up blood draw would be needed, consistent with current clinical guidelines,” she added.

The American Diabetes Association and World Health Organization recommend basing a diabetes diagnosis on at least 2 tests of the 3 standard glycemia measurements—HbA1C levels, fasting glucose level, or 2-hour plasma glucose level after an oral glucose tolerance test—from multiple samples. The organizations encourage repeat tests over a short duration, and in the cases of discordant tests, the positive one should be replicated upon the next patient visit.

According to Selvin, the use of 2 tests from a single blood sample would “potentially [allow for] a major simplification of current clinical practice guidelines.” Physicians already conduct tests like these together, she said. For example, if a patient is obese or has other risk factors for diabetes, the physician is likely to order tests for both fasting glucose and HbA1c from a single blood sample.

“In some sense, our message is an intuitive one,” Selvin said. “It’s just that the guidelines don’t clearly let you use the tests from that 1 blood sample to make the initial diabetes diagnosis.”

K.M. Venkat Narayan, MD, MSc, MBA, and Ram Jagannathan, PhD, authors of an accompanying editorial, told MD Mag that “the analysis performed by [Selvin and her team] is really innovative. We were surprised that this ‘novel’ definition—based on 2 glucose abnormalities in the same blood sample—was strongly associated with for future risk of diagnosed diabetes and its complications.”

The prospective cohort study consisted of 978 patients with elevated levels of fasting glucose or HbA1C. The team defined confirmed undiagnosed diabetes as having elevated levels of fasting glucose of ≥7 mmol/L (≥126 mg/dL) and HbA1C of ≥6.5% from a single blood sample.

In total, 39% of the 978 patients had both elevated HbA1C and fasting glucose (confirmed undiagnosed diabetes) and 61% had 1 elevated measurement (unconfirmed undiagnosed diabetes).

At 15 years, the positive predictive value of the single-sample test was 88.7%, compared with 71.1% for the unconfirmed diabetes cases. In cases where diabetes was confirmed as undiagnosed, there was a strong association with cardiovascular and kidney disease as well as mortality. This association was also stronger than with unconfirmed cases.

The authors did note a single limitation, the lack of repeated measurements of HbA1C and fasting glucose. What this means for physicians, according to Selvin, is that there cannot be direct comparisons of risk associations.

Narayan and Jagannathan said that previous reports have suggested that the fasting plasma glucose and the oral glucose tolerance test (OGTT) have well-known limitations, despite their speed and low cost.

“The observed variability in the course of prediabetes is mostly because of inherent variability in glucose tolerance—particularly as evaluated by the OGTT,” they said. “Although the determination of HbA1c was a better and alternate marker for chronic hyperglycemia, it must be stressed that HbA1c measurement alone is not a sufficiently reliable tool for recognizing particularly the early stages of T2DM or prediabetes.”

For Narayan and Jagannathan, the big take-home message was that even isolated elevation of HbA1c or fasting plasma glucose at a single time point was associated with increased risk of progression to overt diabetes and to complications.

“Therefore, anyone testing abnormal on any glucose test needs to be followed up with a repeat test and also be referred to appropriate lifestyle counseling,” they added. Selvin agreed that conflicting results should be followed up on, which is consistent with the current clinical guidelines.

The study, “Prognostic Implications of Single-Sample Confirmatory Testing for Undiagnosed Diabetes: A Prospective Cohort Study,” was published in the Annals of Internal Medicine.