To determine how survival and clinical status were related to left ventricular (LV) size and systolic function after mitral valve replacement, 104 patients (48 mitral regurgitation [MR], 33 mitral stenosis [MS], and 23 MS/MR) with isolated mitral valve replacement were evaluated before and after surgery. Preoperative hemodynamic abnormalities by cardiac catheterization were improved 6 months after surgery in all three patient groups. The patients with MR exhibited reductions in LV end-diastolic volume index (EDVI) (117 +/- 51 to 89 +/- 27 ml/m2, p less than 0.001) and ejection fraction (EF) (0.56 +/- 0.15 to 0.45 +/- 0.13, p less than 0.001); however, the ratio of forward stroke volume to end-diastolic volume increased (0.32 +/- 0.21 to 0.45 +/- 0.17, p less than 0.001) because of the elimination of regurgitant volume. Survival analysis revealed that mortality was significantly higher in MS or MS/MR patients with postoperative EDVI more than 101 ml/m2 (p less than 0.001 and p less than 0.042, respectively) and in MR patients with postoperative EF less than or equal to 0.50 (p less than 0.031). Also, the majority of patients with MR or MS/MR and postoperative EDVI more than 101 ml/m2 and EF less than or equal to 0.50 were in New York Heart Association class III or IV. Multivariate logistic regression analysis in the patients with MR revealed that the strongest predictor of postoperative EF was preoperative EF (p less than 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)

Because there is little quantitative information about the hemodynamics of left ventricular diastolic events in man, single-plane cineangiographic left ventricular volume curves were quantitatively analyzed at 16.6 msec intervals to evaluate the rate and amount of left ventricular filling during the early passive stage of filling and during the time of atrial contraction. The peak rate of passive diastolic filling of the left ventricle (D dV/dt), the peak rate of filling during atrial contraction (dV/dt c 'a'), and the increment in diastolic left ventricular volume due to atrial contraction (ΔV c 'a') were analyzed for a group of 110 adult patients with a variety of cardiac lesions. Normal D dV/dt, 503 ± 171 cc/sec, was significantly depressed in patients with mitral stenosis (393 ± 109 cc/sec) and coronary artery disease (394 ± 150 cc/sec). When corrected for end-diastolic volume (EDV), the resultant D dV/dt/EDV was significantly depressed in all disease states studied. D dV/dt corrected similarly for total left ventricular stroke volume (SV) was depressed from normal in patients with coronary disease, mitral stenosis, and valvular regurgitation. These data indicate that patients with coronary disease and chronic valvular disease have an abnormality in early diastolic filling of the left ventricle and that this may have a more profound effect on ventricular performance than changes in late diastolic compliance. D dV/dt has a high correlation with peak systolic ejection rate ( r = 0.71), SV ( r = 0.82), and EDV ( r = 0.51). The normal ΔV c 'a' represents 21% of the stroke volume (16% of the stroke volume if corrected for passive filling during this time period). Patients with coronary artery disease, valvular regurgitation, and aortic stenosis have an increased ΔV c'a'. Atrial contraction results in a second peak in mitral valve flow in late diastole (dV/dt c 'a'). In normal patients dV/dt c 'a' is 38% of D dV/dt, but in disease states this ratio is substantially increased. In the diseased heart atrial contraction makes a greater contribution to cardiac output than in the normal heart.

Late follow-up (average = 7.2 years) has been obtained in 249 patients with mitral valve disease who had quantitative angiographic assessment of left ventricular function at thetime of initial catheterization in the 1960s. Surgically treated patients with mitral valve disease had significantly improved survival as compared to medically treated patients with mitral disease. The subgroup with mixed mitral stenosis and regurgitation and the subgroup with moderate impairment of ejection fraction account for this improved survival in surgically treated patients, which occurred despite greater functional and hemodynamic impairment in the surgical cohorts. Using univariate life table survival analysis, ten variables were found to be predictive of survival in the medical cohort, and three in the surgical cohort. With multivariate Cox's regression analysis, end-diastolic volume and arteriovenous oxygen difference were significantly predictive of survival in the medical cohort; age was predictive of survival in the surgical cohort.

The constant (C= 3I) in the hydraulicformulafor calculation of mitral valve area was originally empirically determined by comparison of measured valve area at operation or necropsy with right heart catheterization data. This constant corrects, among other things, for overestimation of the diastolic filling and mitral valve gradient derived from a peripheral arterial tracing and pulmonary capillary wedge pressure, respectively. With left heart catheterization these measurements can be made directly and more accurately. In this study through simultaneous measurement of pulmonary capillary wedge, peripheral arterial, left atrial, and left ventricular pressures, a new constant (K= 40) was derived which should provide a more accurate estimation of mitral valve orifice area when left heart data are used.The hydraulic formula for the calculation of mitral valve area (MVA) developed by Gorlin and Gorlin (I95I) contains a constant (C) empirically calculated by comparison of mitral valve orifice area estimated at operation in ii patients with cardiac output (CO), pulmonary capillary wedge pressure (PC), and diastolic filling period measured peripherally from the brachial artery (pDFP). Left ventricular diastolic pressure was assumed to average 5 mmHg, yielding a mitral valve gradient of PC-5 and the following formula:where HR is heart rate. The constant, C, includes 5 factors: (i) conversion of pressure from mmHg to cm H20 (I-17), (2) deviation of the mitral valve from the 'perfect orifice' for which the formula is valid, (3) overestimation of the actual diastolic filling period (cDFP) because pDFP includes part of the isovolumic contraction and relaxation time, (4) With the development of techniques for retrograde and transseptal catheterization of the left side of the heart, the pressure gradient across the mitral valve (LA-LV) and cDFP are commonly measured directly from left atrial (LA) and left ventricular (LV) pressure tracings. The constant, C, originally estimated for use with indirect measurements will, when used with left heart data, result in an overestimation of mitral valve orifice area. It was the purpose of this study to arrive at a new value for this constant for use with left heart data. For clarity this new constant will be referred to as K, yielding the formula:Methods One way to arrive at a value for K would be to repeat the study of Gorlin and Gorlin (I95I) comparing mitral valve orifice area estimated at operation or measured at necropsy with calculated orifice area from left heart catheterization data obtained shortly before. However, due to the inaccuracies of measuring the mitral orifice at operation and the paucity of intact necropsy specimens, this method was deemed impractical. For the purposes of this study, it was assumed that the value for C of 3I for use with PC-s and pDFP was correct. Since, regardless of the technique of measurement, the mitral orifice is the same; then equation (i) equals equation (2) in a given patient. Solving for K, the following formula is derived:

In a Veterans Administration Cooperative Study involving 13 medical centers, 575 patients undergoing single valve replacement were prospectively randomized to receive either the standard Björk-Shiley prosthesis or the Hancock porcine heterograft (with a modified orifice for sizes 23 and smaller). The hemodynamic data in the 268 patients who underwent cardiac catheterization an average of 6 months (range 3 to 12) postoperatively are reported. Statistical analyses were performed on valve sizes 23, 25 and 27 in the aortic position, and 29, 31 and 33 in the mitral position. A wide variation was observed in mean pressure gradient and calculated orifice area in both valve types within all sizes in both the aortic and the mitral positions. In the aortic position, the Björk-Shiley prosthesis tended to have a lower pressure gradient and larger calculated orifice area than the Hancock heterograft, but the differences in gradient between the two valve types were significant only in the larger-sized valves. The difference in calculated area between the two valve types was not significant within each valve size. In the mitral position, there were no differences in gradient and calculated orifice area between the two types of prostheses. The postoperative cardiac index, regurgitant volume, pulmonary artery systolic and mean pressures, left ventricular end-diastolic pressure, left ventricular ejection fraction and left ventricular end-diastolic volume index did not differ in patients receiving the Björk-Shiley prosthesis from values in patients receiving the Hancock heterograft. Hence, the overall hemodynamic performance of both types of valves is remarkably similar. The choice between these two prostheses should, therefore, be governed not by the hemodynamic performance, but by other factors such as valve durability, risk of anticoagulation and incidence of valve-related complications.