- Email: [email protected]
Echocardiographic determination of left atrial function and its application for assessment of mitral flow velocity pattern
International Journal of Cardiology 72 (1999) 19–25 www.elsevier.com / locate / ijcard
Echocardiographic determination of left atrial function and its application for assessment of mitral flow velocity pattern Guican Zhang M.D., Yoshio Yasumura M.D., Masaaki Uematsu M.D., Satoshi Nakatani M.D., Noritoshi Nagaya M.D., Kunio Miyatake M.D., Masakazu Yamagishi M.D.* Cardiology Division of Medicine, National Cardiovascular Center, 5 -7 -1 Fujishiro-dai, Suita, Osaka 565 -8565, Japan Received 8 February 1999; received in revised form 1 July 1999; accepted 2 August 1999
Abstract We determined left atrial (LA) volume changes to evaluate LA function, and to correlate the Doppler-determined mitral flow velocity (MFV) pattern. Twenty-four patients with ischemic heart disease who showed ‘normal’ MFV pattern by pulsed Doppler echocardiography were studied. The patients were divided into 14 patients with left ventricular end diastolic pressure ,18 mmHg (true normals) and 10 patients with $18 mmHg (pseudo normals). The changes in LA volume were determined by echocardiography from apical two- and four-chamber views with modified Simpson’s method. The volume measurements were done at the time of mitral valve opening (Vmax ), at onset of atrial systole (Va ) and at mitral valve closure (Vmin ). Then the passive LA emptying volume was calculated by subtracting Va from Vmax , and the active LA emptying volume by subtracting Vmin from Va . The LA ejection fraction was calculated by the formula: h(Va 2Vmin )Va j3100. There was no significant difference in LA ejection fraction in pseudo normal (3966%) and in true normal (41613%) patients. Although the passive LA emptying volume was 1664 ml / beat in true normal and was 1163 ml / beat in pseudo normal (NS), the active LA emptying volume was significantly greater in pseudo normals (2264 m / beat) than in true normals (1262 ml / beat, P,0.001). Thus, the ratio of passive and active LA emptying volume was markedly greater in true normals (1.2860.35) than in pseudo normals (0.5260.19, P,0.001), facilitating the differentiation of these two groups. These results indicate that two-dimensional echocardiographic measurement of LA volume can be valuable in assessing the LA function, providing an alternative method for differentiating pseudo normal from true normal MFV pattern in clinical settings, although several technological shortcomings should be resolved. 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Transmitral flow velocity; Heart failure; Left atrial volume
1. Introduction The mitral flow velocity (MFV) pattern determined by conventional Doppler echocardiography has been widely used to estimate the left ventricular diastolic function [1–3]. The normal MFV pattern consists of large rapid filling and small atrial contraction waves [1]. However, this MFV pattern can be affected by several factors such as age [4], heart rate [5], loading *Corresponding author. Tel.: 81-6-6833-5012; fax: 81-6-6833-9865. E-mail address: [email protected] (M. Yamagishi)
conditions [6–8], and left atrial (LA) function [9,10]. Therefore, interpretation of the flow velocity pattern sometimes encounters difficulty, particularly in the presence of the ‘pseudo’ normal MFV pattern that rather indicates a deterioration of the left ventricular diastolic function associated with the elevated left ventricular end diastolic pressure [11]. Based upon the hypothesis that the increase in LA afterload and preload is accompanied by increased LA size and contraction contributing to the determination of MFV pattern, determination of LA volume change may be helpful for the evaluation of pseudo-
0167-5273 / 99 / $ – see front matter 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 99 )00137-0
20
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25
normalization of MFV pattern. Therefore, in the present study we attempted to (1) noninvasively determine the LA volume by echocardiography, and (2) evaluate the indices of LA function for the differentiation of pseudo normal from true normal MFV pattern in clinical settings.
2. Methods
2.1. Study patients From September 1997 to February 1998, we analyzed the MFV pattern and the changes in LA volume in 24 patients with coronary artery disease who underwent coronary angiography and left ventriculography. These patients were selected from a larger number of patients examined according to the following criteria: (1) sinus rhythm; (2) ‘normal’ MFV pattern which was defined as the ratio of rapid filling and atrial contraction waves to be between 1.0 and 2.0; (3) optimal images of LA borders from apical four- and two-chamber views; (4) absence of mitral stenosis nor severe mitral regurgitation. There were 14 patients with angina pectoris and 10 with old myocardial infarction with an average age of 5768 years (range, 37–69 years). All patients underwent echocardiographic examination within 24 h after cardiac catheterization. Under these conditions, there were no significant medical change and hemodynamic state change between noninvasive and invasive studies, although there has been a report suggesting the change of left ventricular pressure determined under a different situation [12]. The catheterization was performed at the absorptive condition without the specific premedication. After diagnostic coronary angiography when the hemodynamics became stable, the left ventricular pressure was measured by fluid-filled manometer that was connected with pressure-recording system. During spontaneous expiration period, the left ventricular pressure was recorded on a computer as well as strip chart system that ran at a paper speed of 100 mm / s. We initially referred the data of the left ventricular pressure from computer, and attempted to confirm the data by manually tracing the original strip chart. The patients were divided into two groups, consisting of 14 patients with true normal MFV with left ventricu-
lar end diastolic pressure ,18 mmHg and 10 patients with pseudo normal MFV ($18 mmHg).
2.2. Echocardiographic examination We used a commercially available echocardiographic equipment with an ultrasound transducer operating at 2.5 MHz (Sonos 2500 or 5500, HewlettPackard). The standard apical four- and two-chamber views in left lateral recumbent position was used to obtain the optimal LA image. The LA images were considered adequate for analysis when at least 75% of the LA border could be traced. Using the loop system, the apical four- and two-chamber views on the same timing were built into the same frame and recorded. The wave form of LA volume change during left ventricular diastolic filling consists of (1) passive empty phase that begins with the opening of the mitral valve to the beginning of atrial diastasis, (2) atrial diastasis phase when the atrial volume remains relatively constant from the beginning of horizontal volume curve to the onset of active atrial systole and (3) active emptying phase that begins with the onset of atrial systole to the minimal atrial volume at the end diastole of the left ventricle (Fig. 1). Therefore, by advancing the frame, the maximal LA volume was measured at the time of mitral valve opening, the diastasis LA volume at the time just before the onset of P wave in ECG, and the minimal LA volume at the time of mitral valve closure. As described by Kircher et al. [13] and Gutman et al. [14], the outline of the LA border was traced along the atrial septum, the posterior left atrial wall and across the lateral LA wall and mitral annulus using a calibrated light pen computational system. When dropout of the atrial septum occurred, or the orifice of the pulmonary veins was seen, the outline was completed by a straight line drawn between the two inner borders of the adjacent clear echoes. Care was taken to exclude both the pulmonary veins and the atrial appendage. The biplane apical four- and two-chamber views were used to measure LA volumes based on biplane modified Simpson’s method with the formula [15]: L(p / 4) o(A i 3Bi ), where L represents the length of the LA, A i and Bi represents the ith disc area of the LA (Fig. 2). Volume measurements were repeated for three times in a single
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25
21
volume was positioned at the level of mitral leaflets from the apical four-chamber view. Then, peak flow velocity of rapid filling and during atrial contraction, and the ratio of these velocities were calculated.
2.3. Indices of left atrial function
Fig. 1. The left atrial (LA) volume change curve determined by automatic boundary detection technique. The wave form of the LA volume curve during left ventricular filling consists of a passive filling phase, a diastasis phase and an active filling phase. The maximal LA volume was measured at the time of mitral valve opening; the LA volume of the diastasis phase was measured at the time just before P wave in ECG; and the minimal LA volume was measured at the time of mitral valve closure.
cardiac cycle and averaged from the images of three consecutive beats. The MFV profile was determined by the conventional pulsed Doppler method in which the sample
For the assessment of LA function, the following indices were determined. The LA passive emptying volume was calculated by subtracting Va from Vmax , the active emptying volume was calculated by subtracting Vmin from Va and the ratio of the LA passive and active emptying volume was computed by the formula: h(maximal LA volume–diastasis LA volume) /(diastasis LA volume2minimal LA volume)j. The LA ejection fraction was calculated by the formula: h(diastasis LA volume2minimal LA volume) / diastasis LA volumej3100. Although there could be other variables in combination LA volume change, we compared above five variables because of the their independence.
2.4. Inter- and intraobserver variabilities To determine inter- and intraobserver variabilities for echocardiographic LA volume measurements, 10 randomly selected measurements were repeated in
Fig. 2. Measurement of left atrial volume by echocardiography from apical (left) four- and (right) two-chamber views with the modified Simpson’s method. A i , Bi , disc area of this slice; LV, left ventricle; RA, right atrium.
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25
22
blinded manner by a second observer (interobserver), and later date second time measurements by the first observer (intraobserver) without reference to previous results from the stored frames of recording videotapes.
2.5. Statistical analysis Results are expressed as mean6SD. Analysis was performed using the unpaired Students’ t-test to compare the pseudo normal and true normal group. Statistical significance was considered as P,0.05.
3. Results
3.1. Patient profiles The patient’s age, body surface area, heart rate and left ventricular stroke volume were not significantly different between pseudo normal and true normal flow pattern patients. Although left ventricular ejection fraction in true normal patients was slightly greater than that in pseudo normal patients (5567 vs. 44616%, P,0.05), LA ejection fraction was not significantly different between two groups (41613 vs. 3966%, NS) (Table 1).
ml, respectively. The maximal LA volume and diastasis LA volume in true normal patients were 4569 and 2965 ml (P,0.001), respectively, and these values were significantly lower than those in pseudo normal patients (P,0.001). The minimal LA volume in pseudo normal patients (34616 ml) was also significantly different from that in true normal patients (1864 ml, P,0.001). Although there was no statistical significance between passive emptying volume of pseudo normal (1163 ml) and that in true normal (1664 ml, NS) patients, the active emptying volume was significantly higher in pseudo normal (2264 ml) than that in true normal (1262 ml) patients (P,0.001). Thus, the ratio of the passive emptying and active emptying volume was markedly greater in true normal (1.2860.35) than in pseudo normal (0.5260.19, P, 0.001) patients (Fig. 3). When the ratio was cut off at 0.9, the sensitivity and specificity for the differentiation of pseudo normal from true normal MFV were 93 and 100%, respectively, facilitating the differentiation of ‘pseudo’ from ‘true’ normal MFV pattern.
3.3. Inter- and intraobserver variabilities
The maximal LA volume and diastasis LA volume in pseudo normal patients were 66617 and 55616
As for the interobserver variability, the measured volumes by each examiner were 44617 and 45619 ml, respectively, yielding a variation of 1064% ( y5 0.92x12.5, r50.93). As for the intraobserver variability, the measured volumes at the different times were 42620 and 43620 ml, yielding the variation of 763% ( y50.95x12.1, r50.96).
Table 1 Patient profiles a
4. Discussion
3.2. LA volume measurements
Age (years) BSA (m 2 ) HR (bpm) LVEDP (mmHg) E /A LVSV (ml / beat) LVEF (%) LAEF (%)
True normal MFV (n514)
Pseudo normal MFV (n510)
P value
5769 1.6760.13 5969 1162 1.2060.20 62613 5567 41613
5968 1.7060.11 5769 2365 1.3860.28 65615 44616 3966
NS NS NS ,0.001 NS NS ,0.05 NS
a A, atrial contraction wave velocity; BSA, body surface area; E, rapid filling wave velocity; HR, heart rate; LVEDP, left ventricular end diastolic pressure; LAEF, left atrial ejection fraction; LVEF, left ventricular ejection fraction; LVSV, left ventricular stroke volume; MFV, mitral flow velocity
4.1. Previous studies The normal MFV pattern in young adults consists of relatively large rapid filling and small atrial contraction waves, thus yielding the ratio of these two waves to be between 1.0 and 2.0 [1]. However, even in the presence of impaired left ventricular function, this MFV pattern maintains ‘normal’ being recognized as a pseudo normal pattern. Under these conditions, it is somewhat difficult to differentiate ‘pseudo’ from ‘true’ normal pattern by conventional Doppler method [16,17]. Although measurements of
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25
23
Fig. 3. Difference in indices of echocardiographic left atrial function in the presence of ‘true’ and ‘pseudo’ normal mitral flow velocity pattern. (left) Although the passive emptying volume was not statistically different in both groups, (center) the active emptying volume and (right) the calculated ratio of the passive and active emptying volumes were significantly different in both groups.
other parameters, such as isovolumic relaxation time, deceleration time of the rapid filling wave, and pulmonary venous flow velocity profiles, may be helpful for differentiation, there existed data showing the sensitivity of these indexes were not satisfactory [18–20].
4.2. Advantage of present study The present data showed that pseudo normal MFV pattern was associated with larger Vmax and Va when compared with those in true normal MFV. The pseudo normal MFV was also associated with a slightly depressed passive LA emptying volume and a significantly increased active LA emptying volume. Therefore, from these variables it is possible to distinguish ‘pseudo’ from ‘true’ normal MFV. Although sensitivity and specificity for differentiation of these MFV patterns seemed to be moderate in the measurements of LA volume, such as maximal LA volume and diastasis LA volume, the ratio of LA passive emptying volume to active emptying volume provided an excellent differentiation, with 93% sensitivity and 100% specificity when the cut-off value was set at 0.9. Therefore, the determination of the ratio is likely to provide an alternative way to differentiate pseudo normal from true normal MFV. Since the LA is considered to serve not only as a reservoir for blood collection, but also a contractile
pump, the evaluation of the LA function is of great importance in the normal as well as in the diseased heart [21–24]. Invasive studies indicated that the LA contracts in a manner similar to the Frank–Starling mechanism [25]. Thus, the LA volume before atrial contraction (diastasis LA volume), which corresponds to the preload of the LA, has a certain relation with LA stroke volume. Indeed, the contribution of the active atrial emptying to the left ventricular filling was reported to increase in patients with myocardial infarction or impaired left ventricular function [21,26]. In the presence of pseudo normal MFV pattern, the LA pressure, as well as the left ventricular end diastolic pressure, increased as also shown in the present data. The increased LA pressure, together with elevated left ventricular end diastolic pressure, might certainly alter LA volume change and also affect LA performance, which resulted in the appearance of pseudo normal MFV pattern.
4.3. Clinical implications Because of complex determinants of MFV profile, it is somewhat difficult to differentiate a pseudo normal from a normal MFV using conventional Doppler echocardiographic variables. For this point of view, the parameter of the ratio of LA passive emptying to active emptying volume may be an
24
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25
alternative index for the clinical differentiation of pseudo from true normal MFV pattern. Application of the on-line measurement of the LA chamber volume using the automatic boundary detection technique may facilitate the measurement of LA volume change in real time (Fig. 1) [27], although we used off-line measurements of LA volume that were relatively complicated. The ratio of passive and active emptying volume of the LA was applied for the diagnosis of pseudonormalization of transmitral flow. From a physiological point of view, it is possible to consider that the decrease in the passive emptying volume is associated with markedly elevated left ventricular end diastolic pressure, and the increase in active emptying volume is associated with a compensatory increase in LA contraction. Under the condition where the left ventricular end diastolic pressure was markedly elevated, the increased LA contraction resulted in an increase in the reflux flow into the pulmonary vein [20].
4.4. Limitations There remain several limitations. First, a lack of simultaneous measurement of left ventricular pressure and LA volume might result in a different condition of the heart in each patient, although the volume measurement was performed within 24 h after catheterization [12]. Blood pressure and heart rate did not significantly change during two measurements, suggesting that the hemodynamic situation was not significantly different in each examination. As for the reproducibility of measurements of LA volume over time in the same subject, this was confirmed as intraobserver variability. Second, it is sometimes difficult to obtain clear LA borders from both apical four- chamber and twochamber views. Therefore, underestimation could occur in the measurement of true LA volumes by conventional echocardiography [13]. However, the use of the volume change ratio as an index for LA function may minimize the effect of underestimation of the absolute volume. Under these conditions, intravenous injection of a contrast agent may facilitate the tracing of the LA border, although we did not use the contrast-enhanced border detection method. Because we did not determine the profile of pulmonary vein flow, which is an alternative index
for determining the left ventricular end diastolic pressure [28], it is somewhat difficult to exactly interpret the relationship between atrial volume change and mitral flow velocity. However, it is possible that retrograde flow to the pulmonary vein contributes to atrial volume change in patients with highly elevated left ventricular end diastolic pressure. Since this was a feasibility study to examine the value of LA volume measurement in the evaluation of MFV, we examined the relatively small number of patients with ischemic heart disease in which the LA was not greatly enlarged. From a clinical point of view, it is important to examine patients with left ventricular dysfunction associated with other diseases, such as cardiomyopathy. Under these conditions, the measurement of enlarged LA volume may be somewhat difficult using the present method. Again, we would suggest that on-line measurement of the LA chamber volume using the automatic boundary detection technique may facilitate the measurement of LA volume change in real time [27].
5. Conclusions The present data indicate that two-dimensional echocardiographic measurement of LA volume is feasible for assessing LA performance. We suggest that this method can be an alternative way to differentiate the pseudo normal from true normal MFV in patients with ‘normal’ transmitral flow velocity pattern, although other Doppler indices such as pulmonary vein flow pattern should also be tried for simultaneous determination.
Acknowledgements Dr Zhang participated in this program with support of a grant from the Japan–China Medical Association. This work was also supported in part by a grant from the Ministry of Health and Welfare (to M.Y.), Tokyo.
References [1] Cohen GI, Pietrolungo IF, Thomas JD, Klein AL. A practical guide to assessment of ventricular diastolic function Doppler echocardiography. J Am Coll Cardiol 1996;27:1753–60.
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25 [2] Appleton CP, Hatle LK, Popp RI. Relation of transmitral flow velocity patterns to left ventricular diastolic function: new insights from a combined hemodynamic and Doppler echocardiographic study. J Am Coll Cardiol 1988;12:426–40. [3] Nishimura RA, Abel MD, Hatle LK, Tajik AJ. Assessment of diastolic function of the heart: background and current applications of Doppler echocardiography. Mayo Clin Proc 1989;64:181–204. [4] Miyatake K, Okamoto M, Kinoshita N et al. Augmentation of atrial contribution to left ventricular inflow with aging as assessed by intracardiac Doppler flowmetry. Am J Cardiol 1984;53:586–9. [5] Harrison MR, Clifton GD, Pennell AT, DeMaria AN. Effect of heart rate on left ventricular diastolic transmitral flow velocity patterns assessed by Doppler echocardiography in normal subjects. Am J Cardiol 1991;67:622–7. [6] Choong CY, Abascal VM, Thomas JD, Guerrero JL, McGlew S, Weyman AE. Combined influence of ventricular loading and relaxation on the transmitral flow velocity profile in dogs measured by Doppler echocardiography. Circulation 1988;78:672–83. [7] Stoddard MF, Pearson AC, Kern MJ, Ratcliff J, Mrosek DG, Labovitz AJ. Influence of alteration in preload on the pattern of left ventricular diastolic filling as assessed by Doppler echocardiography in humans. Circulation 1989;79:1226–36. [8] Nishimura RA, Abel MD, Housmans PR, Warnes CA, Tajik AJ. Mitral flow velocity curves as a function of different loading conditions: evaluation by intraoperative transesophageal Doppler echocardiography. J Am Soc Echocardiogr 1989;2:79–87. [9] Nakatani S, Beppu S, Miyatake K, Nimura Y. Left ventricular function and relationship between left atrial pressure and peak early diastolic filling velocity in dog. Cardiovasc Res 1992;26:109–14. [10] Appleton CP, Galloway JM, Gonzalez MS, Gaballa M, Basnight MA. Estimation of left ventricular filling pressures using twodimensional and Doppler echocardiography in adult patients with cardiac disease. Additional value of analyzing left atrial size, left atrial ejection fraction and the difference in duration of pulmonary venous and mitral flow velocity at atrial contraction. J Am Coll Cardiol 1993;22:1972–82. [11] Nishimura RA, Tajik AJ. Evaluation of diastolic filling of left ventricular in health and disease: Doppler echocardiography is the clinician’s Rosetta Stone. J Am Coll Cardiol 1997;30:8–18. [12] McAnulty JH, Kremkau EL, Rosch J, Hattenhauer MT, Rahimtoola SH. Spontaneous changes in left ventricular function between sequential studies. Am J Cardiol 1974;34:23–8. [13] Kircher B, Abbott JA, Pau S et al. Left atrial volume determination by biplane two-dimensional echocardiography: validation by cine computed tomography. Am Heart J 1991;12:864–71. [14] Gutman J, Wang YS, Wahr D, Schiller NB. Normal left atrial function determined by 2-dimensional echocardiography. Am J Cardiol 1983;51:336–40. [15] Schiller NB, Shah PM, Crawford M et al. Recommendations for quantification of the left ventricle by two-dimensional echocardiography. J Am Soc Echocardiogr 1989;2:362–70.
25
[16] Kuecherer HE, Muhiudeen IA, Kusumoto FM et al. Estimation of mean left atrial pressure from transesophageal pulsed Doppler echocardiography of pulmonary venous flow. Circulation 1990;82:1127–39. [17] Thomas JD, Choong CY, Flachskampf FA, Weyman AE. Analysis of the early transmitral Doppler velocity curve: effect of primary physiologic changes and compensatory preload adjustment. J Am Coll Cardiol 1990;16:644–55. [18] Imanishi T, Nakatani S, Yamagishi M, Beppu S, Miyatake K. Identification of pseudonormal transmitral flow pattern using color Doppler echocardiography. Jpn Circ J 1996;60:749–57. [19] Hashimoto T, Ohte N, Narita H, Kobayashi K, Fujinami T. Differentiation between ‘pseudonormal’ from normal transmitral flow velocity waveforms by evaluating isovolumic relaxation time. J Cardiol 1992;22:523–9. [20] Basnight MA, Gonzalez MS, Kershenovish SC, Appleton CP. Pulmonary venous flow velocity: relation to hemodynamics, mitral flow velocity and left atrial volume and ejection fraction. J Am Soc Echocardiogr 1991;4:547–58. [21] Matsuda Y, Toma Y, Ogawa H et al. Importance of left atrial function in patients with myocardial infarction. Circulation 1983;67:566–71. [22] Toma Y, Matsuda Y, Moritani K, Ogawa H, Matsuzaki M, Kusukawa R. Left atrial filling in normal human subjects: relation between left atrial contraction and left atrial early filling. Cardiovasc Res 1987;21:255–9. [23] Kono T, Sabbah HN, Rosman H, Alam M, Stein PD, Goldstein S. Left atrial contribution to ventricular filling during the course of evolving heart failure. Circulation 1992;86:1317–22. [24] Triposkiadis F, Tentolouris K, Androulakis A et al. Left atrial mechanical function in the healthy elderly: new insights from a combined assessment of changes in atrial volume and transmitral flow velocity. J Am Soc Echocardiogr 1995;8:801–9. [25] Yamaguchi M, Arakawa M, Tanaka T, Takaya T, Nagano T, Hirakawa S. Study on left atrial contractile performance-participation of Frank-Starling mechanism. Jpn Circ J 1987;51:1001–9. [26] Yamasaki T, Takeuchi M, Fujitani K, Fukuzaki H. Effects of left ventricular dysfunction on left atrial performance in previous myocardial infarction and during pacing-induced myocardial ischemia in angina pectoris. Jpn Circ J 1988;52:1240–8. [27] Zhang GC, Tsukada T, Nakatani S et al. Comparison of automatic boundary detection and manual tracing technique in echocardiographic determination of left atrial volume. Jpn Circ J 1998;62:755– 9. [28] Nakatani S, Yoshitomi H, Wada K, Beppu S, Nagata S, Miyatake K. Noninvasive estimation of left ventricular end-diastolic pressure using transthoracic Doppler-determined pulmonary venous atrial flow reversal. Am J Cardiol 1994;73:1017–8.
Echocardiographic determination of left atrial function and its application for assessment of mitral flow velocity pattern Guican Zhang M.D., Yoshio Yasumura M.D., Masaaki Uematsu M.D., Satoshi Nakatani M.D., Noritoshi Nagaya M.D., Kunio Miyatake M.D., Masakazu Yamagishi M.D.* Cardiology Division of Medicine, National Cardiovascular Center, 5 -7 -1 Fujishiro-dai, Suita, Osaka 565 -8565, Japan Received 8 February 1999; received in revised form 1 July 1999; accepted 2 August 1999
Abstract We determined left atrial (LA) volume changes to evaluate LA function, and to correlate the Doppler-determined mitral flow velocity (MFV) pattern. Twenty-four patients with ischemic heart disease who showed ‘normal’ MFV pattern by pulsed Doppler echocardiography were studied. The patients were divided into 14 patients with left ventricular end diastolic pressure ,18 mmHg (true normals) and 10 patients with $18 mmHg (pseudo normals). The changes in LA volume were determined by echocardiography from apical two- and four-chamber views with modified Simpson’s method. The volume measurements were done at the time of mitral valve opening (Vmax ), at onset of atrial systole (Va ) and at mitral valve closure (Vmin ). Then the passive LA emptying volume was calculated by subtracting Va from Vmax , and the active LA emptying volume by subtracting Vmin from Va . The LA ejection fraction was calculated by the formula: h(Va 2Vmin )Va j3100. There was no significant difference in LA ejection fraction in pseudo normal (3966%) and in true normal (41613%) patients. Although the passive LA emptying volume was 1664 ml / beat in true normal and was 1163 ml / beat in pseudo normal (NS), the active LA emptying volume was significantly greater in pseudo normals (2264 m / beat) than in true normals (1262 ml / beat, P,0.001). Thus, the ratio of passive and active LA emptying volume was markedly greater in true normals (1.2860.35) than in pseudo normals (0.5260.19, P,0.001), facilitating the differentiation of these two groups. These results indicate that two-dimensional echocardiographic measurement of LA volume can be valuable in assessing the LA function, providing an alternative method for differentiating pseudo normal from true normal MFV pattern in clinical settings, although several technological shortcomings should be resolved. 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Transmitral flow velocity; Heart failure; Left atrial volume
1. Introduction The mitral flow velocity (MFV) pattern determined by conventional Doppler echocardiography has been widely used to estimate the left ventricular diastolic function [1–3]. The normal MFV pattern consists of large rapid filling and small atrial contraction waves [1]. However, this MFV pattern can be affected by several factors such as age [4], heart rate [5], loading *Corresponding author. Tel.: 81-6-6833-5012; fax: 81-6-6833-9865. E-mail address: [email protected] (M. Yamagishi)
conditions [6–8], and left atrial (LA) function [9,10]. Therefore, interpretation of the flow velocity pattern sometimes encounters difficulty, particularly in the presence of the ‘pseudo’ normal MFV pattern that rather indicates a deterioration of the left ventricular diastolic function associated with the elevated left ventricular end diastolic pressure [11]. Based upon the hypothesis that the increase in LA afterload and preload is accompanied by increased LA size and contraction contributing to the determination of MFV pattern, determination of LA volume change may be helpful for the evaluation of pseudo-
0167-5273 / 99 / $ – see front matter 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 99 )00137-0
20
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25
normalization of MFV pattern. Therefore, in the present study we attempted to (1) noninvasively determine the LA volume by echocardiography, and (2) evaluate the indices of LA function for the differentiation of pseudo normal from true normal MFV pattern in clinical settings.
2. Methods
2.1. Study patients From September 1997 to February 1998, we analyzed the MFV pattern and the changes in LA volume in 24 patients with coronary artery disease who underwent coronary angiography and left ventriculography. These patients were selected from a larger number of patients examined according to the following criteria: (1) sinus rhythm; (2) ‘normal’ MFV pattern which was defined as the ratio of rapid filling and atrial contraction waves to be between 1.0 and 2.0; (3) optimal images of LA borders from apical four- and two-chamber views; (4) absence of mitral stenosis nor severe mitral regurgitation. There were 14 patients with angina pectoris and 10 with old myocardial infarction with an average age of 5768 years (range, 37–69 years). All patients underwent echocardiographic examination within 24 h after cardiac catheterization. Under these conditions, there were no significant medical change and hemodynamic state change between noninvasive and invasive studies, although there has been a report suggesting the change of left ventricular pressure determined under a different situation [12]. The catheterization was performed at the absorptive condition without the specific premedication. After diagnostic coronary angiography when the hemodynamics became stable, the left ventricular pressure was measured by fluid-filled manometer that was connected with pressure-recording system. During spontaneous expiration period, the left ventricular pressure was recorded on a computer as well as strip chart system that ran at a paper speed of 100 mm / s. We initially referred the data of the left ventricular pressure from computer, and attempted to confirm the data by manually tracing the original strip chart. The patients were divided into two groups, consisting of 14 patients with true normal MFV with left ventricu-
lar end diastolic pressure ,18 mmHg and 10 patients with pseudo normal MFV ($18 mmHg).
2.2. Echocardiographic examination We used a commercially available echocardiographic equipment with an ultrasound transducer operating at 2.5 MHz (Sonos 2500 or 5500, HewlettPackard). The standard apical four- and two-chamber views in left lateral recumbent position was used to obtain the optimal LA image. The LA images were considered adequate for analysis when at least 75% of the LA border could be traced. Using the loop system, the apical four- and two-chamber views on the same timing were built into the same frame and recorded. The wave form of LA volume change during left ventricular diastolic filling consists of (1) passive empty phase that begins with the opening of the mitral valve to the beginning of atrial diastasis, (2) atrial diastasis phase when the atrial volume remains relatively constant from the beginning of horizontal volume curve to the onset of active atrial systole and (3) active emptying phase that begins with the onset of atrial systole to the minimal atrial volume at the end diastole of the left ventricle (Fig. 1). Therefore, by advancing the frame, the maximal LA volume was measured at the time of mitral valve opening, the diastasis LA volume at the time just before the onset of P wave in ECG, and the minimal LA volume at the time of mitral valve closure. As described by Kircher et al. [13] and Gutman et al. [14], the outline of the LA border was traced along the atrial septum, the posterior left atrial wall and across the lateral LA wall and mitral annulus using a calibrated light pen computational system. When dropout of the atrial septum occurred, or the orifice of the pulmonary veins was seen, the outline was completed by a straight line drawn between the two inner borders of the adjacent clear echoes. Care was taken to exclude both the pulmonary veins and the atrial appendage. The biplane apical four- and two-chamber views were used to measure LA volumes based on biplane modified Simpson’s method with the formula [15]: L(p / 4) o(A i 3Bi ), where L represents the length of the LA, A i and Bi represents the ith disc area of the LA (Fig. 2). Volume measurements were repeated for three times in a single
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25
21
volume was positioned at the level of mitral leaflets from the apical four-chamber view. Then, peak flow velocity of rapid filling and during atrial contraction, and the ratio of these velocities were calculated.
2.3. Indices of left atrial function
Fig. 1. The left atrial (LA) volume change curve determined by automatic boundary detection technique. The wave form of the LA volume curve during left ventricular filling consists of a passive filling phase, a diastasis phase and an active filling phase. The maximal LA volume was measured at the time of mitral valve opening; the LA volume of the diastasis phase was measured at the time just before P wave in ECG; and the minimal LA volume was measured at the time of mitral valve closure.
cardiac cycle and averaged from the images of three consecutive beats. The MFV profile was determined by the conventional pulsed Doppler method in which the sample
For the assessment of LA function, the following indices were determined. The LA passive emptying volume was calculated by subtracting Va from Vmax , the active emptying volume was calculated by subtracting Vmin from Va and the ratio of the LA passive and active emptying volume was computed by the formula: h(maximal LA volume–diastasis LA volume) /(diastasis LA volume2minimal LA volume)j. The LA ejection fraction was calculated by the formula: h(diastasis LA volume2minimal LA volume) / diastasis LA volumej3100. Although there could be other variables in combination LA volume change, we compared above five variables because of the their independence.
2.4. Inter- and intraobserver variabilities To determine inter- and intraobserver variabilities for echocardiographic LA volume measurements, 10 randomly selected measurements were repeated in
Fig. 2. Measurement of left atrial volume by echocardiography from apical (left) four- and (right) two-chamber views with the modified Simpson’s method. A i , Bi , disc area of this slice; LV, left ventricle; RA, right atrium.
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25
22
blinded manner by a second observer (interobserver), and later date second time measurements by the first observer (intraobserver) without reference to previous results from the stored frames of recording videotapes.
2.5. Statistical analysis Results are expressed as mean6SD. Analysis was performed using the unpaired Students’ t-test to compare the pseudo normal and true normal group. Statistical significance was considered as P,0.05.
3. Results
3.1. Patient profiles The patient’s age, body surface area, heart rate and left ventricular stroke volume were not significantly different between pseudo normal and true normal flow pattern patients. Although left ventricular ejection fraction in true normal patients was slightly greater than that in pseudo normal patients (5567 vs. 44616%, P,0.05), LA ejection fraction was not significantly different between two groups (41613 vs. 3966%, NS) (Table 1).
ml, respectively. The maximal LA volume and diastasis LA volume in true normal patients were 4569 and 2965 ml (P,0.001), respectively, and these values were significantly lower than those in pseudo normal patients (P,0.001). The minimal LA volume in pseudo normal patients (34616 ml) was also significantly different from that in true normal patients (1864 ml, P,0.001). Although there was no statistical significance between passive emptying volume of pseudo normal (1163 ml) and that in true normal (1664 ml, NS) patients, the active emptying volume was significantly higher in pseudo normal (2264 ml) than that in true normal (1262 ml) patients (P,0.001). Thus, the ratio of the passive emptying and active emptying volume was markedly greater in true normal (1.2860.35) than in pseudo normal (0.5260.19, P, 0.001) patients (Fig. 3). When the ratio was cut off at 0.9, the sensitivity and specificity for the differentiation of pseudo normal from true normal MFV were 93 and 100%, respectively, facilitating the differentiation of ‘pseudo’ from ‘true’ normal MFV pattern.
3.3. Inter- and intraobserver variabilities
The maximal LA volume and diastasis LA volume in pseudo normal patients were 66617 and 55616
As for the interobserver variability, the measured volumes by each examiner were 44617 and 45619 ml, respectively, yielding a variation of 1064% ( y5 0.92x12.5, r50.93). As for the intraobserver variability, the measured volumes at the different times were 42620 and 43620 ml, yielding the variation of 763% ( y50.95x12.1, r50.96).
Table 1 Patient profiles a
4. Discussion
3.2. LA volume measurements
Age (years) BSA (m 2 ) HR (bpm) LVEDP (mmHg) E /A LVSV (ml / beat) LVEF (%) LAEF (%)
True normal MFV (n514)
Pseudo normal MFV (n510)
P value
5769 1.6760.13 5969 1162 1.2060.20 62613 5567 41613
5968 1.7060.11 5769 2365 1.3860.28 65615 44616 3966
NS NS NS ,0.001 NS NS ,0.05 NS
a A, atrial contraction wave velocity; BSA, body surface area; E, rapid filling wave velocity; HR, heart rate; LVEDP, left ventricular end diastolic pressure; LAEF, left atrial ejection fraction; LVEF, left ventricular ejection fraction; LVSV, left ventricular stroke volume; MFV, mitral flow velocity
4.1. Previous studies The normal MFV pattern in young adults consists of relatively large rapid filling and small atrial contraction waves, thus yielding the ratio of these two waves to be between 1.0 and 2.0 [1]. However, even in the presence of impaired left ventricular function, this MFV pattern maintains ‘normal’ being recognized as a pseudo normal pattern. Under these conditions, it is somewhat difficult to differentiate ‘pseudo’ from ‘true’ normal pattern by conventional Doppler method [16,17]. Although measurements of
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25
23
Fig. 3. Difference in indices of echocardiographic left atrial function in the presence of ‘true’ and ‘pseudo’ normal mitral flow velocity pattern. (left) Although the passive emptying volume was not statistically different in both groups, (center) the active emptying volume and (right) the calculated ratio of the passive and active emptying volumes were significantly different in both groups.
other parameters, such as isovolumic relaxation time, deceleration time of the rapid filling wave, and pulmonary venous flow velocity profiles, may be helpful for differentiation, there existed data showing the sensitivity of these indexes were not satisfactory [18–20].
4.2. Advantage of present study The present data showed that pseudo normal MFV pattern was associated with larger Vmax and Va when compared with those in true normal MFV. The pseudo normal MFV was also associated with a slightly depressed passive LA emptying volume and a significantly increased active LA emptying volume. Therefore, from these variables it is possible to distinguish ‘pseudo’ from ‘true’ normal MFV. Although sensitivity and specificity for differentiation of these MFV patterns seemed to be moderate in the measurements of LA volume, such as maximal LA volume and diastasis LA volume, the ratio of LA passive emptying volume to active emptying volume provided an excellent differentiation, with 93% sensitivity and 100% specificity when the cut-off value was set at 0.9. Therefore, the determination of the ratio is likely to provide an alternative way to differentiate pseudo normal from true normal MFV. Since the LA is considered to serve not only as a reservoir for blood collection, but also a contractile
pump, the evaluation of the LA function is of great importance in the normal as well as in the diseased heart [21–24]. Invasive studies indicated that the LA contracts in a manner similar to the Frank–Starling mechanism [25]. Thus, the LA volume before atrial contraction (diastasis LA volume), which corresponds to the preload of the LA, has a certain relation with LA stroke volume. Indeed, the contribution of the active atrial emptying to the left ventricular filling was reported to increase in patients with myocardial infarction or impaired left ventricular function [21,26]. In the presence of pseudo normal MFV pattern, the LA pressure, as well as the left ventricular end diastolic pressure, increased as also shown in the present data. The increased LA pressure, together with elevated left ventricular end diastolic pressure, might certainly alter LA volume change and also affect LA performance, which resulted in the appearance of pseudo normal MFV pattern.
4.3. Clinical implications Because of complex determinants of MFV profile, it is somewhat difficult to differentiate a pseudo normal from a normal MFV using conventional Doppler echocardiographic variables. For this point of view, the parameter of the ratio of LA passive emptying to active emptying volume may be an
24
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25
alternative index for the clinical differentiation of pseudo from true normal MFV pattern. Application of the on-line measurement of the LA chamber volume using the automatic boundary detection technique may facilitate the measurement of LA volume change in real time (Fig. 1) [27], although we used off-line measurements of LA volume that were relatively complicated. The ratio of passive and active emptying volume of the LA was applied for the diagnosis of pseudonormalization of transmitral flow. From a physiological point of view, it is possible to consider that the decrease in the passive emptying volume is associated with markedly elevated left ventricular end diastolic pressure, and the increase in active emptying volume is associated with a compensatory increase in LA contraction. Under the condition where the left ventricular end diastolic pressure was markedly elevated, the increased LA contraction resulted in an increase in the reflux flow into the pulmonary vein [20].
4.4. Limitations There remain several limitations. First, a lack of simultaneous measurement of left ventricular pressure and LA volume might result in a different condition of the heart in each patient, although the volume measurement was performed within 24 h after catheterization [12]. Blood pressure and heart rate did not significantly change during two measurements, suggesting that the hemodynamic situation was not significantly different in each examination. As for the reproducibility of measurements of LA volume over time in the same subject, this was confirmed as intraobserver variability. Second, it is sometimes difficult to obtain clear LA borders from both apical four- chamber and twochamber views. Therefore, underestimation could occur in the measurement of true LA volumes by conventional echocardiography [13]. However, the use of the volume change ratio as an index for LA function may minimize the effect of underestimation of the absolute volume. Under these conditions, intravenous injection of a contrast agent may facilitate the tracing of the LA border, although we did not use the contrast-enhanced border detection method. Because we did not determine the profile of pulmonary vein flow, which is an alternative index
for determining the left ventricular end diastolic pressure [28], it is somewhat difficult to exactly interpret the relationship between atrial volume change and mitral flow velocity. However, it is possible that retrograde flow to the pulmonary vein contributes to atrial volume change in patients with highly elevated left ventricular end diastolic pressure. Since this was a feasibility study to examine the value of LA volume measurement in the evaluation of MFV, we examined the relatively small number of patients with ischemic heart disease in which the LA was not greatly enlarged. From a clinical point of view, it is important to examine patients with left ventricular dysfunction associated with other diseases, such as cardiomyopathy. Under these conditions, the measurement of enlarged LA volume may be somewhat difficult using the present method. Again, we would suggest that on-line measurement of the LA chamber volume using the automatic boundary detection technique may facilitate the measurement of LA volume change in real time [27].
5. Conclusions The present data indicate that two-dimensional echocardiographic measurement of LA volume is feasible for assessing LA performance. We suggest that this method can be an alternative way to differentiate the pseudo normal from true normal MFV in patients with ‘normal’ transmitral flow velocity pattern, although other Doppler indices such as pulmonary vein flow pattern should also be tried for simultaneous determination.
Acknowledgements Dr Zhang participated in this program with support of a grant from the Japan–China Medical Association. This work was also supported in part by a grant from the Ministry of Health and Welfare (to M.Y.), Tokyo.
References [1] Cohen GI, Pietrolungo IF, Thomas JD, Klein AL. A practical guide to assessment of ventricular diastolic function Doppler echocardiography. J Am Coll Cardiol 1996;27:1753–60.
G. Zhang et al. / International Journal of Cardiology 72 (1999) 19 – 25 [2] Appleton CP, Hatle LK, Popp RI. Relation of transmitral flow velocity patterns to left ventricular diastolic function: new insights from a combined hemodynamic and Doppler echocardiographic study. J Am Coll Cardiol 1988;12:426–40. [3] Nishimura RA, Abel MD, Hatle LK, Tajik AJ. Assessment of diastolic function of the heart: background and current applications of Doppler echocardiography. Mayo Clin Proc 1989;64:181–204. [4] Miyatake K, Okamoto M, Kinoshita N et al. Augmentation of atrial contribution to left ventricular inflow with aging as assessed by intracardiac Doppler flowmetry. Am J Cardiol 1984;53:586–9. [5] Harrison MR, Clifton GD, Pennell AT, DeMaria AN. Effect of heart rate on left ventricular diastolic transmitral flow velocity patterns assessed by Doppler echocardiography in normal subjects. Am J Cardiol 1991;67:622–7. [6] Choong CY, Abascal VM, Thomas JD, Guerrero JL, McGlew S, Weyman AE. Combined influence of ventricular loading and relaxation on the transmitral flow velocity profile in dogs measured by Doppler echocardiography. Circulation 1988;78:672–83. [7] Stoddard MF, Pearson AC, Kern MJ, Ratcliff J, Mrosek DG, Labovitz AJ. Influence of alteration in preload on the pattern of left ventricular diastolic filling as assessed by Doppler echocardiography in humans. Circulation 1989;79:1226–36. [8] Nishimura RA, Abel MD, Housmans PR, Warnes CA, Tajik AJ. Mitral flow velocity curves as a function of different loading conditions: evaluation by intraoperative transesophageal Doppler echocardiography. J Am Soc Echocardiogr 1989;2:79–87. [9] Nakatani S, Beppu S, Miyatake K, Nimura Y. Left ventricular function and relationship between left atrial pressure and peak early diastolic filling velocity in dog. Cardiovasc Res 1992;26:109–14. [10] Appleton CP, Galloway JM, Gonzalez MS, Gaballa M, Basnight MA. Estimation of left ventricular filling pressures using twodimensional and Doppler echocardiography in adult patients with cardiac disease. Additional value of analyzing left atrial size, left atrial ejection fraction and the difference in duration of pulmonary venous and mitral flow velocity at atrial contraction. J Am Coll Cardiol 1993;22:1972–82. [11] Nishimura RA, Tajik AJ. Evaluation of diastolic filling of left ventricular in health and disease: Doppler echocardiography is the clinician’s Rosetta Stone. J Am Coll Cardiol 1997;30:8–18. [12] McAnulty JH, Kremkau EL, Rosch J, Hattenhauer MT, Rahimtoola SH. Spontaneous changes in left ventricular function between sequential studies. Am J Cardiol 1974;34:23–8. [13] Kircher B, Abbott JA, Pau S et al. Left atrial volume determination by biplane two-dimensional echocardiography: validation by cine computed tomography. Am Heart J 1991;12:864–71. [14] Gutman J, Wang YS, Wahr D, Schiller NB. Normal left atrial function determined by 2-dimensional echocardiography. Am J Cardiol 1983;51:336–40. [15] Schiller NB, Shah PM, Crawford M et al. Recommendations for quantification of the left ventricle by two-dimensional echocardiography. J Am Soc Echocardiogr 1989;2:362–70.
25
[16] Kuecherer HE, Muhiudeen IA, Kusumoto FM et al. Estimation of mean left atrial pressure from transesophageal pulsed Doppler echocardiography of pulmonary venous flow. Circulation 1990;82:1127–39. [17] Thomas JD, Choong CY, Flachskampf FA, Weyman AE. Analysis of the early transmitral Doppler velocity curve: effect of primary physiologic changes and compensatory preload adjustment. J Am Coll Cardiol 1990;16:644–55. [18] Imanishi T, Nakatani S, Yamagishi M, Beppu S, Miyatake K. Identification of pseudonormal transmitral flow pattern using color Doppler echocardiography. Jpn Circ J 1996;60:749–57. [19] Hashimoto T, Ohte N, Narita H, Kobayashi K, Fujinami T. Differentiation between ‘pseudonormal’ from normal transmitral flow velocity waveforms by evaluating isovolumic relaxation time. J Cardiol 1992;22:523–9. [20] Basnight MA, Gonzalez MS, Kershenovish SC, Appleton CP. Pulmonary venous flow velocity: relation to hemodynamics, mitral flow velocity and left atrial volume and ejection fraction. J Am Soc Echocardiogr 1991;4:547–58. [21] Matsuda Y, Toma Y, Ogawa H et al. Importance of left atrial function in patients with myocardial infarction. Circulation 1983;67:566–71. [22] Toma Y, Matsuda Y, Moritani K, Ogawa H, Matsuzaki M, Kusukawa R. Left atrial filling in normal human subjects: relation between left atrial contraction and left atrial early filling. Cardiovasc Res 1987;21:255–9. [23] Kono T, Sabbah HN, Rosman H, Alam M, Stein PD, Goldstein S. Left atrial contribution to ventricular filling during the course of evolving heart failure. Circulation 1992;86:1317–22. [24] Triposkiadis F, Tentolouris K, Androulakis A et al. Left atrial mechanical function in the healthy elderly: new insights from a combined assessment of changes in atrial volume and transmitral flow velocity. J Am Soc Echocardiogr 1995;8:801–9. [25] Yamaguchi M, Arakawa M, Tanaka T, Takaya T, Nagano T, Hirakawa S. Study on left atrial contractile performance-participation of Frank-Starling mechanism. Jpn Circ J 1987;51:1001–9. [26] Yamasaki T, Takeuchi M, Fujitani K, Fukuzaki H. Effects of left ventricular dysfunction on left atrial performance in previous myocardial infarction and during pacing-induced myocardial ischemia in angina pectoris. Jpn Circ J 1988;52:1240–8. [27] Zhang GC, Tsukada T, Nakatani S et al. Comparison of automatic boundary detection and manual tracing technique in echocardiographic determination of left atrial volume. Jpn Circ J 1998;62:755– 9. [28] Nakatani S, Yoshitomi H, Wada K, Beppu S, Nagata S, Miyatake K. Noninvasive estimation of left ventricular end-diastolic pressure using transthoracic Doppler-determined pulmonary venous atrial flow reversal. Am J Cardiol 1994;73:1017–8.