Hemostatic Markers of Endothelial Dysfunction and Risk of Incident Type 2 Diabetes

The Framingham Offspring Study

  1. James B. Meigs1,
  2. Christopher J. O’Donnell2,
  3. Geoffrey H. Tofler3,
  4. Emelia J. Benjamin4,
  5. Caroline S. Fox5,
  6. Izabela Lipinska4,
  7. David M. Nathan6,
  8. Lisa M. Sullivan7,
  9. Ralph B. D’Agostino7 and
  10. Peter W.F. Wilson8
  1. 1General Medicine Division, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts
  2. 2Cardiology Division, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts
  3. 3Cardiology Department, Royal North Shore Hospital, Sydney, New South Wales, Australia
  4. 4Evans Department of Medicine, Preventive Medicine Section, Whitaker Cardiovascular Institute, Boston University School of Medicine, Boston, Massachusetts
  5. 5Division of Endocrinology, Diabetes, and Hypertension, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts
  6. 6Department of Medicine, Diabetes Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts
  7. 7Department of Mathematics, Statistics, and Consulting Unit, Boston University, Boston, Massachusetts
  8. 8Department of Endocrinology, Diabetes, and Medical Genetics, Medical University of South Carolina, Charleston, South Carolina
  1. Address correspondence and reprint requests to James B. Meigs, MD, MPH, General Medicine Division, Massachusetts General Hospital, 50 Staniford St., 9th Floor, Boston, MA 02114. E-mail: jmeigs{at}partners.org


Endothelial dysfunction may precede development of type 2 diabetes. We tested the hypothesis that elevated levels of hemostatic markers of endothelial dysfunction, plasminogen activator inhibitor-1 (PAI-1) antigen, and von Willebrand factor (vWF) antigen predicted incident diabetes independent of other diabetes risk factors. We followed 2,924 Framingham Offspring subjects (54% women, mean age 54 years) without diabetes at baseline (defined by treatment, fasting plasma glucose ≥7 or 2-h postchallenge glucose ≥11.1 mmol/l) over 7 years for new cases of diabetes (treatment or fasting plasma glucose ≥7.0 mmol/l). We used a series of regression models to estimate relative risks for diabetes per interquartile range (IQR) increase in PAI-1 (IQR 16.8 ng/ml) and vWF (IQR 66.8% of control) conditioned on baseline characteristics. Over follow-up, there were 153 new cases of diabetes. Age- and sex-adjusted relative risks of diabetes were 1.55 per IQR for PAI-1 (95% CI 1.41–1.70) and 1.49 for vWF (1.21–1.85). These effects remained after further adjustment for diabetes risk factors (including physical activity; HDL cholesterol, triglyceride, and blood pressure levels; smoking; parental history of diabetes; use of alcohol, nonsteroidal anti-inflammatory drugs, exogenous estrogen, or hypertension therapy; and impaired glucose tolerance), waist circumference, homeostasis model assessment of insulin resistance, and inflammation (assessed by levels of C-reactive protein): the adjusted relative risks were 1.18 per IQR for PAI-1 (1.01–1.37) and 1.39 for vWF (1.09–1.77). We conclude that in this community-based sample, plasma markers of endothelial dysfunction increased risk of incident diabetes independent of other diabetes risk factors including obesity, insulin resistance, and inflammation.


  • P.W.F.W. has been on an ajudication board for Eli Lilly; has received honoraria from Merck and Pfizer; has been a consultant for Eli Lilly, GlaxoSmithKline, and Sanofi; and has received research support from GlaxoSmithKline and Wyeth.

    • Accepted November 9, 2005.
    • Received August 12, 2005.
| Table of Contents