The diastolic stress test (DST) may facilitate the attribution of exertional dyspnoea to cardiac and non-cardiac diseases. However, there is currently no consensus as to the optimal marker of exertional diastolic dysfunction (DD)-the main alternatives being estimated left ventricular (LV) filling pressure (exercise E/e') and diastolic functional reserve (DFRI). We sought to compare the correlates of these parameters.DST was performed by adding the measurement of the transmitral (E) and annular (e') velocities to standard exercise echo in 559 consecutive patients without significant rest or exercise mitral regurgitation. Exertional DD was separately defined by post-stress E/e' >13 or DFRI <13.5. Logistic regression was used to identify the correlates of abnormal responses and linear regression was used to identify the contribution of both to exercise capacity. Abnormal exercise E/e' (n = 112, 20%) and DFRI (n = 317, 57%) were modestly associated (κ 0.35, P < 0.0001). In a linear regression, abnormal exercise E/e' (β = -0.19, P < 0.001) and DFRI (β = -0.15, P = 0.001) were associated with exercise capacity, independent of age, body mass index, wall thickness, haemodynamics or abnormal stress results. Logistic regression revealed abnormal exercise E/e' (R(2)= 0.34) to be independently associated with female gender (β = 0.26, 95% CI: 0.11-0.60, P = 0.002), age (β = 1.04, 95% CI: 1.01-1.07, P = 0.01), hypertension (β = 0.35, 95% CI: 0.15-0.80, P = 0.01) and wall thickness (β = 4.3, 95% CI: 1.3-14.1, P = 0.02). The closest association of abnormal DFRI was exercise capacity (β = 0.89, 95% CI: 0.79-1.02, P = 0.09); no other clinical or stress variable was independently associated.Exercise E/e' and DFRI are both associated with exercise capacity, but E/e' is more closely associated with the expected parameters of DD.