The effect of glucose variability on the long-term risk of microvascular complications in type 1 diabetes

  1. Eric S. Kilpatrick, MD, FRCPath, Honorary Professor in Clinical Biochemistry (Eric.Kilpatrick{at},
  2. Alan S. Rigby, MSc, Senior Lecturer in Statistics and Cardiovascular Epidemiologyb and
  3. Stephen L. Atkin, PhD, FRCP, Professor of Endocrinology and Metabolismc
  1. aDepartment of Clinical Biochemistry, Hull Royal Infirmary, Hull
  2. bAcademic Department of Cardiology, University of Hull and Hull York Medical School, Hull
  3. cDepartment of Diabetes, Hull York Medical School, Hull, UK


    Objective: This study analyzed data from Epidemiology of Diabetes Interventions and Complications (EDIC) study to see if longer term follow-up of Diabetes Control and Complications Trial (DCCT) patients reveals a role for glycemic instability in the development of microvascular complications.

    Research design and methods: The mean area under the curve (AUC) glucose and the within-day glucose variability (SD and MAGE) during the DCCT was assessed to see how they contributed to the risk of retinopathy and nephropathy by year 4 of EDIC.

    Results: Logistic regression showed that the mean glucose during the DCCT and mean HbA1c during EDIC were independently predictive of retinopathy (each p<0.001) and EDIC HbA1c of nephropathy (p=0.001) development by EDIC year 4, whereas glucose variability did not add to this (all p>0.25 using SD or MAGE).

    Conclusions: Glucose variability in the DCCT did not predict the development of retinopathy or nephropathy by EDIC year 4.


      • Received January 20, 2009.
      • Accepted June 14, 2009.