BACKGROUND: Limitations in the predictive value of negative exercise echocardiography in type 2 diabetes mellitus has been linked to a reduced end-systolic pressure-volume response (ESPVR). We sought whether abnormal ESPVR reflected subclinical diabetic heart disease by examining the association between the ESPVR and markers of myocardial dysfunction and to establish if the change (Delta) or peak systolic blood pressure/end-systolic left ventricular volume ratio (SP/ESV) is a better marker of contractile reserve in type 2 diabetes mellitus. METHODS AND RESULTS: Resting and exercise echocardiography was performed in 167 apparently healthy patients with type 2 diabetes mellitus (97 men; age, 55+/-10 years) without ischemia, other cardiac disease, or noncardiac complications of diabetes. Standard echocardiographic and color tissue Doppler measures (early diastolic tissue velocity, strain, and strain rate) were acquired at baseline and peak stress in apical long-axis views. Calibrated integrated backscatter was calculated from a resting parasternal long-axis view. DeltaSP/ESV was calculated as [(peak stress SP/ESV)-(rest SP/ESV)]. The 83 subjects who demonstrated a DeltaSP/ESV < or =12 mm Hg/mL/m(2) after exercise were older and had lower peak heart rate, resting diastolic and stress systolic tissue velocity, stress ejection fraction, and exercise capacity than the remainder. There was no significant association between DeltaSP/ESV and metabolic derangement or echocardiographic measures of deformation or backscatter. Change in Sm and stress ejection fraction were independent correlates of DeltaSP/ESV. CONCLUSIONS: DeltaSP/ESV ratio is associated with established features of subclinical diabetic heart disease as well as determinants of contractile reserve (peak hemodynamic response and stress systolic function). Peak ESPVR is poorly associated with markers of myocardial dysfunction.