This article is more than 5 years old. Information may no longer be current.
Routine blood test may predict diabetes diagnosis
Click Here to Manage Email Alerts
Random plasma glucose tests helped clinicians identify which Veterans Affairs patients should be further tested for diabetes, according to findings recently published in PLoS One.
“Although screening of high-risk patients might be cost-effective, many health care systems have not yet implemented widespread screening,” Mary Rhee, MD, physician-researcher with the Atlanta VA Health Care System and Emory University, told Healio Primary Care.
“Therefore, given the need for convenient, inexpensive ways to identify prediabetes or undiagnosed diabetes, we investigated whether outpatient random plasma glucose levels, frequently included in laboratory tests obtained during routine visits, below diagnostic thresholds (200 mg/dL or higher), could predict subsequent diabetes diagnosis, and thus be used to identify patients who might benefit from more definitive assessment and preventive management,” she said.
Researchers reviewed the records of 942,446 U.S. veterans without diagnosed diabetes, who had three or more random plasma glucose level tests in a baseline year and one or more primary care visit a year during 5 years of follow-up.
They found that 94,599 patients were diagnosed with diabetes during the follow-up period. Those with a diabetes diagnosis had a higher BMI (32 kg/m2 vs. 28 kg/m2), higher random plasma glucose levels (150 mg/dL vs. 107 mg/dL) and were more likely to be black (18% vs. 15%) than those without a diabetes diagnosis. All other clinical characteristics were similar.
Source:Adobe
In addition, having two or more random plasma glucose tests with results showing 115 mg/dL or greater had a specificity of 77% and a sensitivity of 87%. Two or more random plasma glucose tests with results showing 130 mg/dL or greater had a specificity of 93% and a sensitivity of 59%. When only the random plasma glucose test was used to predict diagnosis, receiver operating characteristic was 0.839 within 3 years and 0.803 within 5 years, rates researchers wrote were “reduced but still substantial.”
Rhee discussed how the findings could potentially change clinical practice.
“Use of random plasma glucose levels well below the conventional diagnostic range — 200 mg/dL or higher — obtained opportunistically, during outpatient visits — may signal the need for definitive testing that might lead to earlier identification and permit initiation of preventive management,” she told Healio Primary Care, adding a way to address possible patient pushback to the tests utilized in the study.
“Patients may state that their glucose levels may have been elevated due to a recent snack/meal/drink, and, therefore, they may think that their glucose levels do not reflect their true metabolic state. However, our study specifically investigated random glucose levels, which can reflect glucose levels any time after food ingestion, from immediately afterwards or hours later,” she said in the interview. “Therefore, our findings show that random glucose levels above these thresholds confer diabetes risk regardless of the variability of food/drink ingestion prior to the glucose measurement.”
Rhee added that future studies should perform the same test in nonveteran cohorts to determine the generalizability of the findings. – by Janel Miller
Disclosures: Rhee reports receiving funding from the NIH and the Veterans Administration. Please see the study for all other relevant financial disclosures.