medwireNews: Prediabetes measures are highly variable in their sensitivity and specificity for predicting the long-term risk of diabetes and its major clinical complications, shows an analysis of the Atherosclerosis Risk in Communities (ARIC) cohort.
As reported in The Lancet Diabetes & Endocrinology, Elizabeth Selvin (Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA) and colleagues found that the American Diabetes Association (ADA) fasting glucose prediabetes definition (5.6–6.9 mmol/L) was the most sensitive of the five they studied.
It was 73% sensitive for incident diabetes over 10 years, and was between 46% and 54% sensitive for the major complications of chronic kidney disease, atherosclerotic cardiovascular disease, peripheral vascular disease, and all-cause mortality in 10,844 participants who attended visit 2 of the ARIC study.
However, prediabetes according to this definition was present in 38% of the patients, and, in an accompanying commentary, Lydia Makaroff (University of Leuven, Belgium) notes that “more than one in three people in this population would be diagnosed with a medical condition with the ADA fasting glucose definition, despite their relatively low risk of morbidity and mortality.”
Prediabetes according to the World Health Organization (WHO) fasting glucose definition, of 6.1–6.9 mmol/L, was present in fewer participants, at 11%, but was much less sensitive, at 41% for incident diabetes and 15–19% for major complications. However, it had good specificity, at 91% and 88–89%, respectively.
Both of the glycated hemoglobin (HbA1c) definitions studied had similar specificity to each other and to the WHO fasting glucose definition for incident diabetes and complications. Neither had good sensitivity, however, although the ADA definition of HbA1c cutoff (5.7–6.4%) was more sensitive than the narrower International Expert Committee definition (6.0–6.4%).
The team also tested the predictive ability of the ADA and WHO 2-hour glucose concentration prediabetes definition (7.8–11.0 mmol/L) among 7194 participants who attended study visit 4, but did not find any “meaningful differences” in its predictive ability compared with the fasting glucose definitions.
“The choice of which test and threshold to use to identify people with intermediate hyperglycaemia will depend on the specific needs and goals of the screening programme,” observes Makaroff.
“Researchers will need to find the correct balance between the false positive rate and false negative rate that is acceptable in their particular setting.”
However, she calls for “a more precise term” than prediabetes, given that the five in-use definitions “identify different, though overlapping, groups of people.”
She suggests: “It might be useful to instead use more precise (though less patient-friendly) terms such as impaired glucose tolerance, impaired fasting glucose, and sub-diabetic HbA1c.”
medwireNews is an independent medical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2016