Left ventricular chamber stiffness from model-based image processing of transmitral Doppler E-waves

Background Model-based image processing (MBIP) of Doppler E-waves eliminates the need for digitizing waveforms by hand or determining the contour 'by eye'. Little et al. (Circulation 1995, 92:1933-1939) used pressure-volume measurements for dogs to verify the physiologic-model-derived prediction that the left ventricular chamber stiffness, K-LV, can be determined from the deceleration time t(dec), when that portion of the E-wave contour is fit by a cosine function. MBIP of clinical Doppler E-wave images to determine chamber stiffness K-LV has not been performed. Objective We sought to determine K-LV by MBIP of clinical Doppler E-wave images and elucidate the physiologic meaning of the harmonic oscillator filling model's parameter k. Methods and results The unique mathematical relationship between the kinematic, harmonic oscillator model of filling and K-LV predicts that the oscillator's spring constant k be linearly proportional to the chamber stiffness K-LV. To verify this, digitally acquired, clinical Doppler transmitral flow velocity images from 21 subjects were analyzed. The parameter k and the stiffness K-LV were computed independently for each subject and compared. In accordance with prediction, a linear relationship between k and the stiffness K-LV, namely k = 1.16 [A/(rho L)]K-LV + 41, r = 0.96, was observed. Conclusions The oscillator parameter k is linearly proportional to the left ventricular chamber stiffness K-LV. The MBIP approach allows automated computation of k and K-LV, provides a robust, automated, observer-independent method of Doppler transmitral flow velocity analysis, and eliminates the need for visual determination of the contour or measurement of its attributes by eye. It provides a stimulus for further validation of the relationships among k, K-LV, and catheterization-based diastolic chamber properties in humans and their correlations with selected diastolic function-altering syndromes. (C) Rapid Science Publishers.