The normal human left ventricular response to large variations in preload was studied in 12 young men. M-mode echocardiograms were recorded at supine rest and compared with studies obtained dur-ing head-down tilt (HDT) at 50 and during progressive lower body negative pressure (LBNP) to-40 mm Hg. During HDT, end-diastolic volume (EDV) increased 23%(p< 0.001), stroke volume (SV) increased 35%(p< 0.001) and ejection fraction (EF) increased 10%(p< 0.05). Heart rate (HR) decreased by 5 beats/min (p< 0.025). During LBNP, EDV decreased 28%(p< 0.001), end-systolic volume (ESV) decreased 21%(p< 0.001) and SV decreased 33%(p< 0.001). LBNP was accompanied by a minor increase in HR (9 beats/min, p< 0.001) and a small increase in systolic blood pressure (11 mm Hg, p< 0.01). Comparisons between mea-surements obtained during HDT and LBNP showed significant differences in EDV, ESV, SV, EF, HR and diastolic blood pressure. The mean velocity of circumferential fiber shortening was unchanged. The combined data from the control studies and the interventions defined the normal left ventricular function curve as an exponential function where SV= 0.36 EDVY''ml (r= 0.93, p< 0.001), which was not significantly different from the linear regression SV= 0.6 EDV+ 0.57 (r= 0.92, p< 0.001). Our data indicate that the mean velocity of circumferential fiber shortening is an index of contractile state that is independent of preload, whereas other echocardiographic measurements are significantly altered by large changes in preload.

THE RESPONSE of the normal human left ventricle to changes in preload has beenstudied by various methods. 1 12 However, the range of preload variation has been limited, and the results have often been difficult to interpret due to simultaneous changes in ar-terial blood pressure and heart rate. Previousstudies in our laboratory showed that large variations in pre-load with minimal effects on heart rate or arterial blood pressure can be introduced noninvasively by head-down tilt and lower body negative pressure (LBNP). 12, 13 The primary purpose of the study was to define echocardiographically the extreme ranges of the normal relationship between left ventricular end-diastolic dimensions and stroke volume (ie, a classic Frank-Starling curve). We also studied the effect of wide variations in preload on commonly used indexes of myocardial contractile state.