Endothelial Adhesion Molecules in Diabetes - Adhesion Molecules - page 186

of both atherosclerotic cardiovascular disease (CVD) and type 2 diabetes (Fig. 1) .
Although there is some evidence to link several direct measures of endothelial
dysfunction (such as l ow-mediated dilation of the brachial artery or retinal
arteriolar narrowing) to subsequent risk of type 2 diabetes, the role of elevated
levels of endothelial adhesion molecules in predicting diabetes risk remains
inconclusive because of limited prospective data.
Results from all prospective studies available by February 2009 are summarized
in Table 1. A few prospective studies have directly evaluated the role of endothelial
biomarkers in predicting future risk of developing type 2 diabetes but have
yielded mixed results (Krakof et al. 2003, Meigs et al. 2004, Herder et al. 2006,
h orand et al. 2006, Song et al. 2007). In the i rst nested case-control study of
Pima Indians, none of the endothelial biomarkers, E-selectin, ICAM-1, and
VCAM-1, was signii cantly associated with type 2 diabetes. h e small sample size
of this study may limit its statistical power to detect a signii cant association. By
contrast, in a larger case-control study from the Nurses' Health Study, Meigs et al.
(2004) reported that elevated levels of E-selectin and ICAM-1 were independent
predictors of incident diabetes in initially non-diabetic Caucasian women and
VCAM-1 was not associated with diabetes risk. Further adjustment for baseline
levels of C-reactive protein (CRP), fasting insulin, and hemoglobin A1c did not
change these associations. Consistent with these i ndings, E-selectin but not
ICAM-1 was an independent predictor of type 2 diabetes risk in a population-
based case-cohort study of middle-aged German men and women (h orand et al.
2006). E-selectin remained signii cantly associated with diabetes risk in men and
women at er additionally controlling for age, body mass index, smoking, alcohol,
physical activity, SBP, total cholesterol/HDL cholesterol ratio, parental history of
diabetes, and CRP. In addition, two prospective cohort studies have presented data
on ICAM-1 levels only (Herder et al. 2006, Sattar et al. 2009) and their results have
also been inconsistent (Table 1) .
In a case-control study nested within the Women's Health Initiative
Observational Study (WHI-OS), an ethnically diverse cohort of US post-
menopausal women including whites, blacks, Hispanics, and Asians and Pacii c
Islanders, we found ethnic dif erences in plasma levels of endothelial adhesion
molecules and coni rmed the predictive role of both E-selectin and ICAM-1 in all
ethnic groups (Fig. 2) (Song et al. 2007). When these three endothelial markers
were mutually adjusted, VCAM-1 was no longer predictive of diabetes risk in two
studies, and E-selectin remains the strongest diabetes predictor among these three
biomarkers (Fig. 3) . Elevated levels of endothelial adhesion molecules may to
some extent rel ect a chronic inl ammatory state. However, incremental changes
in circulating levels of E-selectin remain independently associated with future
diabetes risk irrespective of CRP levels at baseline.
Taken together, prospective data appear to support a robust association between
E-selectin and incident diabetes. h ere are several plausible explanations. First,