Use Continuous Glucose Monitoring to Diagnose Diabetes?

TOPLINE:

• Continuous glucose monitoring (CGM) data provide a more comprehensive characterization of glucose values than fasting glucose (FG) measurement alone.
• Considerable FG variability was seen even in the same individual, suggesting CGM can improve diabetes diagnostic precision.

METHODOLOGY:

• Analysis of data from the 10 K study, including FG values from 8315 individuals aged 40-70 years obtained during 59,565 mornings (median, 7 days per participant), plus 2-week CGM data.
• FG values were classified as either normal (< 100 mg/dL [5.6 mmol/L]), prediabetes (100-125 mg/dL [5.6-6.9 mmol/L]), or diabetes (≥ 126 mg/dL [7.0 mmol/L]).

TAKEAWAY:

• Mean overall FG value was 96.2 mg/dL, rising by 0.234 mg/dL with each year of age in women and 0.25 mg/dL in men.
• Of 8044 individuals who had a ≥ 1 valid morning window, the mean standard deviation of FG in the same individual was 7.52 mg/dL.
• Throughout the study, only 46.94% of participants had FG values that stayed consistent by category.
• Among 5328 individuals who would have been classified with normal FG at study initiation, only 57% had all other sequential FG measurements in the normal range, while 40% would have been reclassified as prediabetes and 3% would have been reclassified as either suspected (2%) or diagnosed (1%) diabetes.
• Among 2718 individuals who would have been considered to have prediabetes, 7% would have been reclassified with diabetes and 11% suspected to have diabetes.
• By contrast, 12% with initial prediabetes diagnosis had completely normal FG values during follow-up.

IN PRACTICE:

“Our findings suggest that careful consideration is necessary when interpreting FG as substantial intraperson variability exists and highlight the potential impact of using CGM data to refine glycemic status assessment.”

SOURCE:

This study was conducted by Smadar Shilo, MD, PhD, of Weizmann Institute of Science, Rehovot, Israel, and colleagues. It was published online in Nature Medicine.

LIMITATIONS:

Possible inaccurate morning meal logging by participants. No information on certain factors that could influence glucose levels, such as stress, physical activity, or menstrual cycle. All individuals were 40- to 70-year-old Israelis (including immigrants), generalizability uncertain. Possible CGM accuracy limitations.

DISCLOSURES:

One author received support from the Crown Human Genome Center, the Larson Charitable Foundation New Scientist Fund, the Else Kröner-Fresenius Foundation, the White Rose International Foundation, the Ben B. and Joyce E. Eisenberg Foundation, the Nissenbaum Family Foundation, M. Pinheiro de Andrade and V. Buchheim, M. Michels, and A. Moussaief and grants funded by the Minerva Foundation, with funding from the German Federal Ministry for Education and Research; the European Research Council; and the Israel Science Foundation. Another author was partially supported by the Israeli Council for Higher Education via the Weizmann Data Science Research Center. One author is an employee in Pheno.AI, Ltd, a biomedical data science company from Tel Aviv, Israel, and two others are paid consultants to Pheno.AI, Ltd. The other authors, including Shilo, declared no competing interests.