Increased left atrial size has been identified as a precursor of atrial fibrillation and of stroke once atrial fibrillation is manifest. Conflicting data exist regarding the effect of high blood pressure on left atrial size. Our objective was to evaluate the association of contemporary and long-term measures of blood pressure with echocardiographically determined left atrial size in a large, population-based cohort. The study sample consisted of 1849 male and 2152 female participants of the Framingham Heart Study and Framingham Offspring Study. All analyses were sex specific. In correlation analyses, systolic and pulse pressures were identified as statistically significant determinants of left atrial size after adjustment for age and body mass index, although the magnitudes of these relations were very modest (partial r < or = .10). Multivariable linear regression models showed the relative contributions of the pressure variables to the prediction of left atrial size to be substantially less than those of age and, in particular, body mass index. Furthermore, inclusion of left ventricular mass in these multivariable models eliminated or attenuated the associations of the pressure variables with left atrial size. In logistic analyses, increasing levels of the pressure variables were significantly predictive of left atrial enlargement. Subjects with 8-year average systolic pressure of 140 mm Hg or higher were twice as likely to have left atrial enlargement as those with values of 110 mm Hg or lower. Overall, in this population-based study sample, increased levels of systolic and pulse pressures (but not diastolic or mean arterial pressures) were significantly associated with increased left atrial size.(ABSTRACT TRUNCATED AT 250 WORDS)