The changes of left ventricular diastolic function during exercise are heterogeneous in patients with hypertrophic cardiomyopathy (HCM). We sought to investigate whether exertional changes of mitral annular velocities and plasma N-terminal-pro-B-type natriuretic peptide (BNP) concentration are associated with exercise capacity in patients with HCM. After a comprehensive echocardiographic study, 32 patients with HCM performed symptom-limited graded supine bicycle exercise. Echocardiographic Doppler parameters were measured at each stage of exercise. Blood samples were collected at rest and immediately after exercise to determine the concentration of pro-BNP. Resting pro-BNP (r = -0.620, P < .001), E'(base) (r = 0.414, P = .018), and DeltaE'(50W) (change of E' from base to 50 W of exercise) (r = 0.367, P = .039) were significantly correlated with exercise duration. Left ventricular longitudinal diastolic function reserve index, defined as DeltaE' x E'(base), significantly correlated with exercise duration (at 50 W, r = 0.540, P = .001) independent of age, sex, body mass index, and resting pro-BNP level. When combining DeltaE' x E'(50W) (<5.85 cm(2)/s(2)) and resting pro-BNP (>740 mg/dL), the predictive accuracy for exercise capacity could be improved (for <500 seconds, global chi(2) = 5.84 in pro-BNP vs 8.10 in pro-BNP + DeltaE' x E'(50W), P = .023). The assessment of left ventricular longitudinal diastolic function reserve can provide incremental information to pro-BNP for the prediction of exercise capacity in patients with HCM.