Noninvasive assessment of left ventricular (LV)

Transcription

1 Comparative Value of Tissue Doppler Imaging and M-Mode Color Doppler Mitral Flow Propagation Velocity for the Evaluation of Left Ventricular Filling Pressure* Michal Kidawa, MD; Lisa Coignard, MD; Gérard Drobinski, MD, PhD; Maria Krzeminska-Pakula, MD, PhD; Daniel Thomas, MD; Michel Komajda, MD; and Richard Isnard, MD Background: Recently, two new indexes based on the ratio of transmitral early diastolic velocity (E) to tissue Doppler imaging (TDI), and early diastolic velocity of mitral annulus (E ) and E to propagation velocity (Vp) have been proposed to predict left ventricular (LV) filling pressures. However, little is known about the comparative value of these two indexes. Methods: We studied 71 consecutive patients referred for coronary angiography (mean age SD, years; 21 patients with LV ejection fraction [EF] < 50%). Complete Doppler echocardiographic examination including TDI and Vp measurements and direct measurement of LV end-diastolic pressure (LVEDP) were performed simultaneously in the catheterization laboratory. LV filling pressures were considered elevated when LVEDP was > 15 mm Hg. Results: The correlation coefficients between E/E and E/Vp and LVEDP were 0.68 (p 0.01) and 0.54 (p 0.01), respectively, in the overall population. The correlations were better in patients with low LV EF (< 50%) [0.8 (p 0.01) and 0.77(p 0.01)] and poor in patients with normal LV EF (0.57 [p 0.05] and 0.41 [not significant]), respectively. Moreover, Vp measurements had higher interobserver variability compared to E (14% vs 7%). The cutoff values for both indexes giving the best sensitivity and specificity in identifying LVEDP > 15 mm Hg were 9 for (E/E ) and 2 for (E/Vp) Conclusion: Both E/E and E/Vp can be used for the evaluation of LV filling pressures. However, the sensitivity of these indexes, especially E/Vp, is hampered by EF. E/E has a lower variability than Vp and should be preferred for estimation of filling pressures especially in patients with EF > 50%. (CHEST 2005; 128: ) Key words: diastole; echocardiography; pressure; tissue Doppler Abbreviations: A late diastolic velocity; A late diastolic velocity of the mitral annulus; E early diastolic velocity; E early diastolic velocity of the mitral annulus; EF ejection fraction; IVRT isovolumic relaxation time; LV left ventricular; LVEDP left ventricular end-diastolic pressure; TDI tissue Doppler imaging; Vp propagation velocity Noninvasive assessment of left ventricular (LV) filling pressures is a key issue in clinical practice. Pulsed Doppler echocardiographic measurements of transmitral flow velocities have been shown to be useful in this setting but only in selected patients with reduced LV ejection fraction (EF). 1 5 The early *From the Department of Cardiology (Drs. Kidawa and Krzeminska-Pakula), Medical University of Łódź, Łódź, Poland; and Hôpital Pitié-Salpetrierem (Drs. Coignard, Drobinski, Thomas, Komajda, and Isnard), Institut de Cardiologie, Paris France. This study was supported in part by scientific grant founded by Servier International. Manuscript received March 29, 2005; revision accepted April 5, diastolic velocity (E) and deceleration are actually dependent on multiple interrelated factors, including mainly preload and relaxation. To overcome these limitations, it has been proposed to combine the transmitral flow velocities to other Doppler parameters, including pulmonary venous flow velocities or response of transmitral flow to decreased Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( org/misc/reprints.shtml). Correspondence to: Michal Kidawa, MD, 2nd Chair and Department of Cardiology, Medical University of Łódź, Poland, Biegański Hospital, Ul. Kniaziewicza 1/5, Łódź, Poland; mkidawa@wp.pl, mkidawa@ptkardio.pl 2544 Clinical Investigations

2 loading conditions Recently, tissue Doppler imaging (TDI) of the mitral annulus during diastole and color M-mode derived flow propagation velocity (Vp) have been described as good indicators of ventricular relaxation, with a relative preload independency. Therefore, two new indexes based on the ratio of E to TDI early diastolic velocity of the mitral annulus (E ), and E to Vp have been proposed to predict LV filling pressure However, little is known about which index should be preferred for practical use. The aim of this study was to evaluate and compare accuracy and usefulness of E/Vp and E/E in prediction of high LV filling pressures in patients with normal and low EFs. Patients Materials and Methods We enrolled 88 consecutive patients in sinus rhythm referred for clinically coronary angiography. Patients with acute coronary syndrome, organic mitral or aortic valve disease, or heart transplant recipients were not included. The remaining 71 patients (mean age SD, years; 13 women) formed our study group. All the patients gave informed consent in agreement with ethics regulations. Doppler Echocardiographic Studies All patients were examined in the catheterization laboratory. Simultaneously with invasive pressure recordings, two-dimensional and Doppler echocardiographic examinations were performed with an ultrasonographic system (Sequoia 256; Acuson; Mountain View, CA) equipped with multifrequency transducer. LV EF was calculated from apical two- and four-chamber views using a modified Simpson rule. Transmitral flow patterns were recorded from apical four-chamber windows with 2- to 3-mm pulsed-sample Doppler volume placed between mitral valve tips in diastole during five consecutive cardiac cycles. Maximal velocities of E- and late diastolic velocity (A)-waves, deceleration time of E, A-wave duration time, and isovolumic relaxation time (IVRT) were measured. Pulmonary venous flow was assessed on right upper pulmonary vein, with sample volume positioned 5 to 10 mm proximal to its junction with the left atrium; velocities of the systolic reversal wave, diastolic reversal wave, atrial reversal wave, and pulmonary A-wave duration were measured. The measurement of Vp was performed in apical four-chamber view by color Doppler echocardiography in M-mode. Then, adjustment of Doppler window and Nyquist velocity to two thirds of blood flow peak velocity was done to display the average velocity of mitral early wave from the mitral annulus to 4 cm toward the apex of the left ventricle. Vp of the early wave was measured as the slope of the line parallel to the recorded border between blue and red colors (which illustrates Nyquist velocity). M-mode color and pulsed Doppler signals were recorded at a horizontal sweep of 100 mm/s. Tissue Doppler Measurements The tissue Doppler program was set in pulsed-wave Doppler mode. Motion of mitral annulus was recorded in the apical four-chamber view. Sample volume was positioned sequentially at the lateral and septal corners of the mitral annulus. Two major negative velocities were recorded with the movement of the annulus toward the base of the heart during diastole, as follows: one during the early phase of diastolic myocardial velocity (E ), and another during the late phase of diastolic myocardial velocity (late diastolic velocity of the mitral annulus [A ]). A major positive systolic velocity was recorded with the movement of the annulus toward the cardiac apex during systole. The peak myocardial systolic velocity was defined as the maximum velocity during systole, excluding the isovolumic contraction. All velocities were recorded for five consecutive cardiac cycles, and the results were averaged. TDI measurements of peak E and A were made for each cycle, and the mean was calculated. All tissue Doppler signals were recorded at horizontal time sweep set at 100 mm/s. Pressure Measurements Baseline LV end-diastolic pressure (LVEDP) recordings were acquired before coronary angiography and ventriculography. LV pressure measurements were done invasively with a 7F, fluid-filled pigtail catheter (Cordis Corporation; Miami, FL) with pressure transducers after calibration. The fourth intercostal space in the anterior axillary line was used as the zero level. We recorded LV diastolic pressures as follows: minimal pressure, catheterizationinvestigated pre-a-wave pressure and LVEDP. We defined LVEDP as maximal pressure drop after pressure increase due to atrial contraction and before the rise of systolic pressure. Pressure data were collected at end-expiration. Three consecutive heart cycles were evaluated, and the mean value of LVEDP was calculated. LV filling pressures were considered elevated in case of LVEDP 15 mm Hg. Data Analysis All data acquired during echocardiographic examinations were stored in digital imaging and communications in medicine format on magneto-optic disk. Analysis of obtained echocardiographic recordings and measurements was performed off-line on personal computer (Tomtec Imaging System, Image Arena Version 2.7; Tomtec; Munich, Germany) by two experienced echocardiographists with calculation of absolute differences and variability. Ratios of E/A, E/E, E/Vp, systolic/diastolic reversal waves, and difference between mitral A-wave duration and pulmonary atrial reversal duration (mitral pulmonary A-wave duration) were calculated. For each result, the average taken from at least three recordings was used. Statistical analysis was performed using software (Statistica 6.0; StatSoft; Tulsa, OK). All data are presented as mean and SD. Continuous variables were compared using Student t test for unpaired data when appropriate. Linear regression analysis was performed to evaluate the relationship between echocardiographic variables and invasive pressure measurements. A p value of 0.05 was considered statistically significant. Results Characteristics of the patient study group are presented in Table 1. Diagnosis of coronary artery disease was confirmed in 58 patients, while 6 patients had normal coronary arteries, 7 patients had low EF, and normal coronary arteries were diagnosed as dilated cardiomyopathy. Criteria of elevated LVEDP were fulfilled in group of 42 patients (mean CHEST / 128 / 4/ OCTOBER,

3 Table 1 Baseline Characteristics of the Study Group Characteristics Data Mean age, yr Female/male gender 13 (18)/58 (82) Coronary artery disease 58 (82) Normal coronary arteries 6 (8) Dilated cardiomyopathy 7 (10) New York Heart Association class I/II/III 30 (50)/36 (43)/5 (7) EF 50% 21 (29) Heart rate, beats/min Systolic BP, mm Hg Diastolic BP, mm Hg LVEDP, mm Hg Maximum 37 Minimum 6 LVEDP 15 mm Hg 42 (59) *Data are presented as mean SD or No. (% of patients). LVEDP, mm Hg). Success ratios in obtaining good-quality signals and interobserver reproducibility of selected echocardiographic parameters are presented in Table 2 Comparison of Doppler Echocardiographic Data Within Subgroups Classified on the Basis of LV Filling Pressures and LV EF In patients with high filling pressures with normal and low EFs, both E and E/A were higher, as compared to patients with LVEDP 15 mm Hg (Table 3). Atrial reversal duration and the difference between pulmonary and mitral A-wave durations were higher, whereas diastolic/systolic reversal wave ratio values were lower in patients with high filling pressures and normal LV EF. Vp values were similar in all groups; however, the E/Vp ratio was significantly higher in patients with elevated filling pressures but only in those with low LV EF. While TDI parameters such as lateral E, lateral A, septal E, septal A, and septal E/E were Table 2 Success Ratios in Obtaining Good-Quality Signals and Interobserver Reproducibility of Selected Parameters. Variables Success Ratio, % Absolute Difference, SD Mean Difference, % Systolic reversal wave, m/s Diastolic reversal wave, m/s Atrial reversal wave, m/s Vp, m/s Lateral E, m/s Lateral A, m/s Septal E, m/s Septal A, m/s similar in all groups, only the lateral E/E ratio was significantly higher in patients with LVEDP 15 mm Hg disregarding LV EF. Correlation of Echocardiographic Parameters to LV Filling Pressures E and E/A ratio and mitral A-wave duration correlated significantly with LV filling pressures in patients with EF 50% (Table 4; Fig 1, 2). Similarly, the E/Vp ratio correlated significantly with LVEDP in patients with low LV EF (Fig 1). On the opposite, pulmonary and mitral A-wave durations correlated with LVEDP only in patients with normal LV EF. Lateral or septal E diastolic Doppler tissue velocities did not correlate with LVEDP. In all patients and in those with EF 50%, the lateral E/E ratio presented higher correlation values with filling pressures than septal E/E. In patients with preserved systolic function, only E/E of the lateral wall correlated significantly with LVEDP (Fig 2). Usefulness of Selected Parameters in Identifying Patients With Elevated LV Filling Pressures E/Vp and E/E were tested to determine the best sensitivity and specificity in identifying patients with LVEDP 15 mm Hg (Table 5). In patients with LV EF 50%, E/E 9 predicted LVEDP 15 mm Hg with sensitivity of 86% and specificity of 83%, and 60% and 84%, respectively, in those with normal LV EF; while E/Vp 2 predicted LVEDP 15 mm Hg with a sensitivity of 85% and a specificity of 83%, and 48% and 76%, respectively, in patients with LV EF 50%,. Discussion Doppler echocardiographic assessment of LV filling pressures remains difficult in clinical practice. Recently, two new indexes based on E to TDI E and E to Vp emerged as the most reliable indicators for predicting elevated LV filling pressures in patients with normal and low EF. However, the comparison of the accuracy and usefulness of these two indexes has been poorly investigated. In this study, the E/E ratio appears to be more accurate than E/Vp in estimation of LV filling pressures, especially in patients with normal LV EF. We also found that E measurements were easier to obtain and were more reproducible than Vp. In publications comparing E/E and E/Vp, similar results reporting slight advantage of E/E over E/Vp were reported by Nagueh et al 16 ; however, this study was based on a selected population of patients with hypertrophic cardiomyopathy and thus cannot be 2546 Clinical Investigations

4 Table 3 Summarized Baseline and Echocardiographic Data; Comparison Between Groups Classified According to LV EF and Measurements of Invasive LV Filling Pressures* EF 50% EF 50% Variables LVEDp 15 LVEDp 15 LVEDp 15 LVEDp 15 Patients, No E, m/s A, m/s E/A ratio Deceleration time, ms IVRT, ms Systolic reversal wave, m/s Diastolic reversal wave, m/s Systolic/diastolic reversal wave Pulmonary A-wave duration, ms Mitral pulmonary A-wave duration Vp, m/s E/Vp Lateral E, m/s Lateral A, m/s Lateral E/E Septal E, m/s Septal A, m/s Septal E/E *Data are presented as mean SD. p 0.05, LVEDP 15 mm Hg vs 15 mm Hg. p 0.01, LVEDP 15 mm Hg vs 15 mm Hg. applied to all patients. Results reported by Rivas- Gotz et al 17 also suggested a superiority of E/E over E/Vp on a population comparable to ours. These authors 17 also described the negative influence of Table 4 Correlation of Transmitral Flow Variables and Vp and E/Vp With LVEDP Transmitral Flow All Patients EF 50% EF 50% E, m/s * 0.45 A, m/s E/A ratio * 0.40 Deceleration time, ms IVRT, ms Mitral A-wave duration, ms * 0.13 Vp, m/s E/Vp 0.54* 0.77* 0.41 Systolic reversal wave, m/s Diastolic reversal wave, m/s Systolic/diastolic reversal wave Atrial reversal wave, m/s Pulmonary A-wave 0.57* * duration, ms Mitral pulmonary A-wave 0.43* * duration, ms Lateral E, m/s Lateral A, m/s Lateral E/E 0.68* 0.80* 0.58* Septal E, m/s Septal A, m/s * 0.09 Septal E/E 0.52* 0.72* 0.29 *p high LV systolic function on echocardiographic indexes used for noninvasive LV filling pressure estimation. We also documented a decrease of the relationship between E/E, E/Vp, and LVEDP in patients with high EF. The sensitivity for detecting an elevated LVEDP in patients with normal EF is poor, especially for E/Vp Other studies 11,12 also performed direct comparison of E/E and E/Vp indexes in estimating pulmonary capillary wedge pressure and found better correlation with the E/Vp index than with E/E. However, most of their patients were receiving mechanical ventilation with different underlying diseases, including valvular disease and acute coronary events, and were in ICUs. Several factors might affect the relationship between pulmonary capillary wedge pressure and filling pressures, particularly in critically ill patients with regional variations in pulmonary vasculature, thus making applying these results to a widespread population difficult. Firstenberg et al 14 observed that TDI indexes could be affected by changes in loading conditions, while Vp was not, and that E/E ratio was less accurate than E/Vp in predicting of LV filling pressures. However, these results concerned only healthy volunteers and not patients with underlying cardiac disease. We also observed higher variability in the measurements of Vp as compared to E between two independent observers in agreement with previous studies CHEST / 128 / 4/ OCTOBER,

5 Figure 1. E/Vp, E/E lateral vs LVEDP in patients with EF 50%. Figure 2. E/Vp, E/E lateral vs LVEDP in patients with EF 50% Clinical Investigations

6 Table 5 Sensitivity and Specificity of Various Cutoff Values for Lateral E/E and E/Vp* Variables EF 50% EF 50% Sensitivity Specificity Sensitivity Specificity Lateral E/E Lateral E/E Lateral E/E Lateral E/E Lateral E/E E/Vp E/Vp E/Vp *Data are presented as %. In our study, correlation coefficients of E/E and LVEDP were lower than those reported at first by Nagueh et al 6 but were in agreement with more recent reports. 7,19 We also found that correlation values between LVEDP and E/E ratio were different, depending on the site of the measurement of mitral annulus. Contrary to other studies, 19 E/E ratio derived from TDI of the intraventricular septum presented modest correlation values in the whole study group. We demonstrated that only in patients with decreased systolic function, the septal E/E ratio correlated significantly with LVEDP. A possible explanation of this fact could be a lower success ratio in acquisition of good quality TDI signals from the intraventricular septum, and higher intraobserver variability as compared to TDI of the lateral wall. In addition to the main findings, we observed a significant correlation between mitral pulmonary A-wave duration and LVEDP in patients with an EF 50% with slight superiority over TDI-derived parameters. These results are similar with data presented by Paraskevaidis et al 20 and Poerner et al, 15 who demonstrated that differences in duration of the pulmonary venous retrograde velocity and the transmitral A-wave velocity enabled the most accurate estimation of LVEDP. However, these authors 15 also indicated that TDI-derived indexes E/E were reliable alternatives in the assessment of LVEDP, superior to the index derived from Vp. Although assessment of pulmonary venous flow could be helpful in assessment of filling pressures, these recordings, unlike tissue Doppler images, are sometimes difficult to obtain by the transthoracic window. Limitations We studied clinically stable patients with ischemic heart disease, and therefore our conclusions are limited to this group of patients. The influence of regional wall motion dysfunction on mitral annulus movement is still unknown; therefore, we cannot exclude that it could have affected E/E. In our study, we have limited TDI measurements to two sites the intraventricular septum and the lateral wall and we did not examine anterior and posterior wall velocities that might have revealed additional information. Also, we cannot exclude the possibility that E could be affected by preload changes according to Firstenberg et al. 21 Because of technical limitations in invasive pressure recordings, LVEDP, not catheterization-investigated pre-a-wave pressure, was used as surrogate for LV filling pressures. Conclusions Both E/E and E/Vp can be used for the evaluation of LV filling pressures. However, E/E has a lower variability than Vp and should be preferred for estimation of filling pressures especially in patients with EF 50%. References 1 Yalcin F, Kaftan A, Muderrisoglu H, et al. Is Doppler tissue velocity during early left ventricular filling preload independent? Heart 2002; 87: Hurrell DG, Nishimura RA, Ilstrup DM, et al. Utility of preload alteration in assessment of left ventricular filling pressure by Doppler echocardiography: a simultaneous catheterization and Doppler echocardiographic study. J Am Coll Cardiol 1997; 30: Yamamoto K, Nishimura RA, Chaliki HP, et al. Determination of left ventricular filling pressure by Doppler echocardiography in patients with coronary artery disease: critical role of left ventricular systolic function. J Am Coll Cardiol 1997; 30: Nishimura RA, Appleton CP, Redfield MM, et al. Noninvasive Doppler echocardiographic evaluation of left ventricular filling pressures in patients with cardiomyopathies: a simultaneous Doppler echocardiographic and cardiac catheterization study. J Am Coll Cardiol 1996; 28: Garcia MJ, Thomas JD, Klein AL. New Doppler echocardiographic applications for the study of diastolic function. J Am Coll Cardiol 1998; 32: Nagueh SF, Middleton KJ, Kopelen HA, et al. Doppler tissue imaging: a noninvasive technique for evaluation of left ventricular relaxation and estimation of filling pressures. J Am Coll Cardiol 1997; 30: Nagueh SF, Mikati I, Kopelen HA, et al. Doppler estimation of left ventricular filling pressure in sinus tachycardia: a new application of tissue Doppler imaging. Circulation 1998; 98: Sohn DW, Chai IH, Lee DJ, et al. Assessment of mitral annulus velocity by Doppler tissue imaging in the evaluation of left ventricular diastolic function. J Am Coll Cardiol 1997; 30: Sundereswaran L, Nagueh SF, Vardan S, et al. Estimation of left and right ventricular filling pressures after heart transplantation by tissue Doppler imaging. Am J Cardiol 1998; 82: De Boeck BW, Cramer MJ, Oh JK, et al. Spectral pulsed CHEST / 128 / 4/ OCTOBER,

7 tissue Doppler imaging in diastole: a tool to increase our insight in and assessment of diastolic relaxation of the left ventricle. Am Heart J 2003; 146: Garcia MJ, Ares MA, Asher C, et al. An index of early left ventricular filling that combined with pulsed Doppler peak E velocity may estimate capillary wedge pressure. J Am Coll Cardiol 1997; 29: Gonzalez-Vilchez F, Ayuela J, Ares M, et al. Comparison of Doppler echocardiography, color M-mode Doppler, and Doppler tissue imaging for the estimation of pulmonary capillary wedge pressure. J Am Soc Echocardiogr 2002; 15: Nagueh SF, Kopelen HA, Quinones MA. Assessment of left ventricular filling pressures by Doppler in the presence of atrial fibrillation. Circulation 1996; 94: Firstenberg MS, Levine BD, Garcia MJ, et al. Relationship of echocardiographic indices to pulmonary capillary wedge pressures in healthy volunteers. J Am Coll Cardiol 2000; 36: Poerner TC, Goebel B, Unglaub P, et al. Detection of a pseudonormal mitral inflow pattern: an echocardiographic and tissue Doppler study [abstract]. Echocardiography 2003; 20: Nagueh SF, Lakkis NM, Middleton KJ, et al. Doppler estimation of left ventricular filling pressures in patients with hypertrophic cardiomyopathy. Circulation 1999; 99: Rivas-Gotz C, Manolios M, Thohan V, et al. Impact of left ventricular ejection fraction on estimation of left ventricular filling pressures using tissue Doppler and flow propagation velocity. Am J Cardiol 2003; 91: Seo Y, Ishimitsu T, Moriyama N, et al. Estimating pulmonary capillary wedge pressures using Doppler variables of early diastolic left ventricular inflow. Jpn Circ J 2001; 65: Ommen SR, Nishimura RA, Appleton CP, et al. Clinical utility of Doppler echocardiography and tissue Doppler imaging in the estimation of left ventricular filling pressures: a comparative simultaneous Doppler-catheterization study. Circulation 2000; 102: Paraskevaidis IA, Tsiapras DP, Karavolias GK, et al. Dopplerderived left ventricular end-diastolic pressure prediction model using the combined analysis of mitral and pulmonary A waves in patients with coronary artery disease and preserved left ventricular systolic function. Am J Cardiol 2002; 90: Firstenberg MS, Greenberg NL, Main ML, et al. Determinants of diastolic myocardial tissue Doppler velocities: influences of relaxation and preload. J Appl Physiol 2001; 90: Clinical Investigations