Estimating Pulmonary Capillary Wedge Pressures Using Doppler Variables of Early Diastolic Left Ventricular Inflow

Yoshihiro Seo, Toshiyuki Ishimitsu, Naoko Moriyama, Tomoko Ishizu, Tsuyoshi Enomoto, Michiko Sakane, Hideki Kamiya, Yuri Hiranuma, Iwao Yamaguchi
2001 Japanese Circulation Journal - English Edition  
ulmonary capillary wedge pressure (PCWP) is an important clinical marker of cardiac function, 1-3 but measuring it requires right heart catheterization, which limits longterm monitoring or its use as a screening test. Therefore, a simple, reliable, and noninvasive method of estimating PCWP is required. Previous experimental studies have shown that left atrial pressure (LAP) is the most effective component of the transmitral flow velocity profile (TMF) obtained by pulsed Doppler
more » ... 4-9 and there have been numerous investigations of the relationship between Doppler variables, including the early diastolic transmitral flow peak velocities (E), the late diastolic transmitral flow peak velocities (A), the ratio E to A (E/A), the deceleration time of E (DT) and the isovolumic relaxation time (IVRT), and PCWP or left ventricular end-diastolic pressure (LVEDP). 12-15 Some clinical studies have pointed out, however, that TMF cannot be used in all types of patients to estimate PCWP, because the relationship between PCWP and Doppler variables varies with the subject. 16, 17 Other studies have also suggested that the LAP can be estimated more accurately using combinations of Doppler variables 18-25 ; for example, it has recently been shown that the propagation velocity (PV) of early diastolic left ventricular (LV) inflow based on color Doppler M-mode measurements was found to correlate well with the time constant of isovolumic LV relaxation (tau), and may be used to evaluate LV relaxation. 27, 28, 30, 32 On the basis that E depends on LAP and LV relaxation, 9-11 the ratio of E and PV (E/PV) has been shown to correlate well with PCWP. [23] [24] [25] However, we considered that PCWP might correlate well with multilinear regression analysis using E and PV in the interpretation of the determinant component of E. Accordingly, the present study was designed to evaluate whether a multilinear regression model based on E and PV can estimate PCWP more accurately than other Dopplerbased methods in various cardiac disease states. Methods Study Protocol In the initial study group, Doppler echocardiography and cardiac catheterization were performed simultaneously, and an equation to estimate PCWP was obtained by multilinear regression analysis using E and PV. The reliability of the obtained equation was then tested prospectively in a second population (test population). In this second study, the patients were classified into 3 groups based on LV function and morphology: (1) LV systolic dysfunction, (2) LV hypertrophy, and (3) no evidence of LV systolic dysfunction or hypertrophy. We determined the correlations between PCWP and Doppler TMF variables, and between the estimated PCWP obtained from the regression equation and the PCWP measured using a Swan-Ganz catheter. All patients gave written informed consent before participation. Initial Study Group The initial group comprised 30 patients (21 men, 9 women) whose ages ranged from 35 to 80 years (mean, 57±19 years). Twenty patients underwent catheterization for the evaluation of cardiac function and coronary artery disease: 5 had idiopathic with dilated cardiomyopathy, 3 had hypertrophic cardiomyopathy, and 2 had aortic stenosis. Patients with mitral stenosis, moderate to severe mitral Jpn Circ J 2001; 65: 33 -39 The present study was performed to determine whether a multilineal regression model based on the early diastolic transmitral flow peak velocity (E) and the propagation velocity of early diastolic inflow (PV) could estimate the pulmonary capillary wedge pressure (PCWP). PCWP and Doppler variables were simultaneously recorded in 30 patients. PCWP was estimated by multilinear regression analysis using E and PV. The predictive accuracy of the equation obtained from the analysis was tested prospectively in a separate group of 65 patients divided into 3 groups: left ventricular (LV) systolic dysfunction (Group A), LV hypertrophy (Group B), and preserved systolic function without hypertrophy (Group C). The initial results obtained in groups B and C, respectively, were: r=0.77; r=0.81. These results indicate that a multilinear regression model based on E and PV is a noninvasive method of accurately estimating PCWP in a variety of cardiac disease states. (Jpn Circ J 2001; 65: 33 -39)
doi:10.1253/jcj.65.33 pmid:11153819 fatcat:6lfjvizx6vau7luckhrsbkmfvq