The changes that occur in diastolic indices during normal aging, including reduced velocity of early filling (E), lengthened E deceleration time (DT), augmented late filling (A), and prolonged isovolumic relaxation time (IVRT), have been attributed to a slower rate of left ventricular (LV) pressure decay. Indeed, this constellation of findings is often referred to as “abnormal relaxation” pattern. However, early LV filling is determined by the atrioventricular pressure gradient, which depends on both the LV pressure decline and the left atrial (LA) pressure. IVRT is also highly dependent upon LA pressure. To assess the relative influence of the rate of LV pressure decline and LA pressure on parameters of diastolic filling, and their association with age, we studied 122 normal subjects (aged 21 to 92) in the Baltimore Longitudinal Study of Aging by Doppler echocardiography and cardiac magnetic resonance imaging (MRI). LV pressure decline was assessed using 2 previously validated, load-insensitive methods: filling wave velocity propagation by color M-mode (Vp), and LV untwisting rate (recoil rate) by tagged MRI. Early diastolic LA pressure was evaluated using 4 methods: pulmonary vein flow systolic fraction, pulmonary vein flow diastolic deceleration time, color M-mode (E/Vp), and tissue Doppler echocardiography (E/Em). Standard echocardiographic parameters including E, A, IVRT, and DT were also recorded. Regression of variables against age showed the expected reduction of E (r=0.40, p<0.0001), increase in A (r=0.51, P<0.0001), and prolongation of IVRT (r=0.61, p<0.0001) and DT (r=0.56, p<0.0001), with advancing age. The duration of systole was prolonged (r=0.38, p<0.0007) with aging. There was no relation of age to Vp or recoil rate, parameters reflecting the rate of LV pressure decline. However, older age was associated with reduced E/Vp (r=0.36, p=0.008) and increased pulmonary vein systolic fraction (r=0.58, p<0.001), pulmonary vein deceleration time (r=0.42, p=0.0026), and E/Em (r=0.51, p<0.0001), all suggesting reduced early LA pressure with advancing age. By multiple regression analysis, older age and smaller Vp were independently associated with reduced peak E velocity, but when an index of LA pressure was included in the model, the age effect dropped out. Therefore, the reduced early diastolic LV peak filling rate seen in healthy older adults may be more closely related to a reduced LA pressure during early diastole, than to slower LV pressure decline. This effect would also explain later mitral valve opening, manifesting as prolonged IVRT. We postulate that changes in LA active or passive properties may contribute to the development of the “abnormal relaxation” pattern during the aging process.