This study was designed to answer two questions. First, does the left ventricular contractile dysfunction resulting from mitral regurgitation (MR) reflect a primary defect in the cardiac muscle cell? Second, what is the basis for any change in cellular contractile function that might be observed? Left ventricular volume overload was produced in 10 dogs by catheter transection of mitral chordae tendineae. Three months later in these and in seven control dogs, left ventricular contractile function was characterized by the end-ejection stress-volume relation (EESVR). Investigators who were blinded to these results then characterized the contractile performance of cardiac muscle cells, or cardiocytes, from these same left ventricles in terms of the viscosity (graded external load) -velocity relation. Finally, the tissue and cellular components of these same left ventricles were analyzed morphometrically. Both the left ventricles from the MR group and their constituent cardiocytes showed marked contractile abnormalities. By matching ventricles with cells from the same MR dogs, ventricular EESVR was correlated with cardiocyte peak sarcomere shortening velocity (SSV). The correlation coefficient between EESVR and SSV was 0.63, but between a size-independent measure of active ventricular stiffness and SSV, it was 0.88. No change in left ventricular interstitial volume fraction was found in MR dogs, but both ventricular and cellular contractile dysfunction strongly correlated with a decreased volume fraction of cardiocyte myofibrils. Last, in an attempt to relate the degree of contractile dysfunction to the hypertrophic response, left ventricular mass in the MR dogs was correlated with both cellular and ventricular contractile indexes; no significant correlation was found. Three conclusions are warranted by these studies. First, chronic left ventricular volume overload from mitral regurgitation leads to contractile defects at both the ventricular and cellular levels, the extent of which correlates well in individual animals. Second, no quantitative interstitial change resulted from MR. Taken together, these two findings strongly suggest that the contractile defect is intrinsic to the cardiocyte. Third, while the contractile abnormality in MR remains undefined, the most basic defects appear to be a combination of myofibrillar loss with the failure of compensatory hypertrophy to occur in response to progressive decrements in cellular and ventricular function.