In the March 2003 issue of Arteriosclerosis, Thrombosis, and Vascular Biology, in the article by Keaney et al (Obesity and Systemic Oxidative Stress: Clinical Correlates of Oxidative Stress in The Framingham Study; pp 434–439), there is an error in the legend for the figure. The authors apologize for this error, and the correct legend appears below.
Systemic oxidative stress related to BMI. Nontransformed urinary creatinine–indexed 8-epi-PGF2α (isoprostane) are plotted against BMI quintile for the entire sample*: men (top panel) and women (lower panel). The range of BMI (kg/m2) for each quartile as well as the number of observations are included. Data represent the mean±SEM of each quintile. The relation between 8-epi-PGF2α levels and BMI was significant for both men and women by multivariable linear regression (P<0.001).
*Not for nonsmokers, as originally stated in the legend.