BMC Endocrine Disorders (Feb 2019)

Preventing misdiagnosis of diabetes in the elderly: age-dependent HbA1c reference intervals derived from two population-based study cohorts

  • Annette Masuch,
  • Nele Friedrich,
  • Johannes Roth,
  • Matthias Nauck,
  • Ulrich Alfons Müller,
  • Astrid Petersmann

DOI
https://doi.org/10.1186/s12902-019-0338-7
Journal volume & issue
Vol. 19, no. 1
pp. 1 – 10

Abstract

Read online

Abstract Background Measurement of gylcated hemoglobin A1c (HbA1c) plays a central role in monitoring quality of antidiabetic therapy and in the diagnosis of diabetes. Several studies report increased levels of HbA1c in nondiabetic elderly. However, this observation did not reach incorporation into daily clinical practice or the respective guidelines. The present study aimed to evaluate HbA1c levels in relation to age in two independent population-based cohorts and to derive age-specific reference intervals. Methods Four thousand two hundred sixty three participants from the Study of Health in Pomerania (SHIP-0) and 4402 participants from the independent study SHIP-Trend were included. HbA1c was determined by means of high-performance liquid chromatography. Multivariable linear regression models were performed. Reference intervals for HbA1c were determined. Results Reference intervals were derived from a healthy subpopulation with the upper reference limit (URL) for HbA1c of 42.1 mmol/Mol (6.0%) for individuals aged 20–39 years increasing to 43.2 mmol/Mol (6.1%) for individuals aged 40–59 years. For people aged ≥60 years the URL was 47.5 mmol/Mol (6.5%). In both study populations an increase in HbA1c with age was observed. ANOVA revealed up to 8.5 mmol/Mol (0.77%) or 7.3 mmol/Mol (0.68%) higher estimated mean levels of HbA1c in the oldest compared to the youngest age group in SHIP-0 or SHIP-trend, respectively. Linear regression analyses confirmed the positive associations of HbA1c with age which was independent of BMI Conclusion The present study confirmed the previously observed increase of HbA1c with increasing age in non-diabetic individuals. As a consequence age-dependent reference values for HbA1c were derived from two large and well defined reference populations. Implementation of them into daily practice may improve patient care and diagnosis of diabetes and reduce the risk of misdiagnosis and subsequent overtreatment of diabetes in elderly patients.

Keywords