Effects of posture on left ventricular diastolic filling during exercise

Abstract

SATO, A., A. KOIKE, Y. KOYAMA, T. YAJIMA, F. MARUMO, and M. HIROE. Effects of posture on left ventricular diastolic filling during exercise. Med. Sci. Sports Exerc., Vol. 31, No. 11, pp. 1564–1569, 1999.

Background: 

Measuring the transmitral flow velocity with Doppler echocardiography is a useful method for evaluating left ventricular diastolic function. However, there are few data regarding the effect of posture during exercise on transmitral flow velocity.

Methods: 

The transmitral flow velocity with pulsed-wave Doppler echocardiography was measured during supine and upright bicycle ergometer exercise in 10 normal young men without cardiac disease (26.7 ± 5.5 yr).

Results: 

The ratio of the early rapid filling wave to the atrial filling wave (E/A) was gradually decreased with increasing exercise intensity. At rest and during recovery, the E/A ratio was significantly higher (P < 0.01) in the supine position than in the upright position. This difference was caused mainly by the higher E wave in the supine position. However, E wave and E/A ratio did not differ between the upright and supine position during exercise.

Conclusion: 

Although measurement of left ventricular filling is completely noninvasive and clinically useful for evaluating diastolic function, it was found that the E/A ratio was profoundly influenced by posture and exercise intensity. These results suggest that the potential influences of posture and exercise intensity on the filling velocities should be taken into account when interpreting diastolic function by Doppler echocardiography.

Abnormality in left ventricular diastolic function is an early manifestation of cardiac dysfunction and occurs in a variety of cardiovascular diseases (^10,11). Until the 1980s, clinical methods for assessing diastolic function had been limited to the direct measurement of intracardiac pressure. However, measurement of the transmitral flow velocity with pulsed-wave Doppler echocardiography was found to be useful for evaluating diastolic function (^2,17). Because Doppler echocardiography is completely noninvasive, this method is now in wide use in the clinical setting for estimation of the degree of diastolic dysfunction in cardiac patients (^22,30).

Transmitral flow velocity during the diastolic period consists of early rapid-filling (E) wave and atrial-filling (A) wave. Abnormal relaxation patterns in the filling velocity, which are characterized by a low E/A ratio, have been described in patients with diastolic dysfunctions such as hypertensive heart disease (²⁴), coronary artery disease (^9,36), and hypertrophic cardiomyopathy (^20,34).

Previous studies have demonstrated that Doppler-derived indexes of left ventricular diastolic function are affected by loading conditions (^5,6,13,32), heart rate (^7,8,12), systolic function (³), and the subject’s age (^21,31). However, there are no data regarding the effects of posture on transmitral flow velocity. Postural change may profoundly affect stroke volume through the Frank-Starling mechanism (²⁵). If so, it could also have influence on diastolic function, which is reflected in the filling velocity. These influences may be more apparent during exercise than during the resting state.

In this study, we compared the change of transmitral filling velocity between supine and upright ergometer exercise in normal subjects. The purpose of the study was to clarify the effects of posture on the left ventricular filling and to ascertain whether this effect is pronounced or attenuated during exercise.

METHODS

Subjects.

Ten normal young men without cardiac or pulmonary disease were investigated (Table 1). Each subject voluntarily consented to participate in the study. The protocol and procedures for this study were approved by the institution’s human subjects committee. The nature of the study, its purposes, and risks were explained to each subject, and written informed consent was obtained before enrollment.

Table 1: Physical characteristics and echocardiographic findings obtained at rest in the lateral decubitus position for each subject.

Exercise protocol.

Exercise tests were employed using an electromagnetically braked cycle ergometer (WLP-400, Lode, Holland). On the study day, each subject performed incremental exercise tests both in the supine and upright position in random order: five subjects performed the upright exercise first, and the remaining five subjects performed the supine exercise first. The interval between two tests was approximately 30 min. After 2 min of rest, the exercise test began with a 3-min warm-up at 20 W and 60 rpm, and the load was then increased incrementally by 1 W every 3 s. For the first test, exercise was continued as long as the E and A waves of transmitral flow could be visually separated and was terminated just before they overlapped each other. For the second test, exercise was terminated immediately when the subject reached the maximal work rate of the first test.

The heart rate was continuously monitored using a Case II Stress System (Marquette Electronics, Milwaukee, WI). Cuff blood pressures were taken every minute with an automatic manometer (STBP-680F, Collin Denshi, Aichi, Japan) (¹⁶).

Echocardiographic and Doppler recordings.

All examinations were performed with a Hewlett-Packard Sonos 1000 Ultrasound Device (Hewlett Packard Co., Andover, MA) equipped with a 2.5-MHz probe. Two-dimensional echocardiography was performed at rest in the lateral decubitus position. Left ventricular diameters and wall thickness were measured by M-mode echocardiography (²⁷). Ejection fraction was measured by two-dimensional echocardiography by application of an area-length method to an apical four-chamber view, according to American Society of echocardiography recommendations (²⁸). Left ventricular inflow velocities were obtained from the apical four-chamber view by using pulsed-wave Doppler echocardiography. The sample volume was placed at the leaflet tips of the mitral valve with the ultrasound beam oriented parallel to the left ventricular inflow. The transmitral flow was continuously recorded on videotape from the beginning of the 2-min resting period preceding exercise through 10 min of recovery after exercise. All recordings were taken by the same echocardiographer who is highly experienced in Doppler measurement. The data was also analyzed by an experienced observer who was unaware of the subjects’ clinical data.

Respiratory gas measurements.

Breath-by-breath oxygen uptake was measured throughout the test through 10 min of recovery by using a Respiromonitor RM-300 (Minato Medical Science, Osaka, Japan) (^14,29).

This system consists of a microcomputer, a hot-wire flowmeter, and a gas analyzer comprising a sampling tube, filter, suction pump, zirconium-element oxygen analyzer, and infrared carbon-dioxide analyzer. Gas was drawn by the suction pump and sampled at a rate of 220 mL·min⁻¹ through the filter in the analyzers. The system was carefully calibrated before each study.

Data analysis.

The heart rate and oxygen uptake were calculated every minute. It has been reported that oxygen pulse calculated by dividing the oxygen uptake by the heart rate correlates well with the stroke volume determined by the direct Fick method during exercise. Thus, we measured oxygen pulse (oxygen uptake/heart rate) as a representative of stroke volume. Pulsed-wave Doppler images of the transmitral velocity profile were analyzed by digitalizing the profile and averaging values from three consecutive cycles. The peak velocities of A and E waves and the ratio of E to A waves were calculated every minute (Fig. 1).

Figure 1: Pulsed Doppler recording of mitral valve flow at rest in a normal subject. A, atrial-filling wave velocity; E, early rapid-filling wave velocity.

Statistics.

Data are reported as mean ± SD. Differences in the variables at rest and at peak exercise between supine and upright positions were analyzed by a Student’s paired t-test. Differences in the time course of variables during exercise and recovery between supine and upright positions were analyzed by a two-way repeated measures ANOVA. Differences were considered statistically significant when p was < 0.05.

RESULTS

Physical characteristics and echocardiographic findings are shown in Table 1 for each subject. Left ventricular diastolic dimension, systolic dimension, and ejection fraction at rest were 45 ± 2 mm, 27 ± 3 mm, and 71 ± 4%, respectively. By study design, the maximal work rate attained during upright and supine exercise, 98 ± 20 W, was exactly the same for each subject (Table 2).

Table 2: Heart rate at maximal exercise and maximal work rate obtained during incremental exercise for each subject.

Heart rate.

The heart rate at rest and during exercise and recovery is shown in Figure 2A. In the upright position, the heart rate was 80 ± 10 bpm at rest and increased to 117 ± 11 bpm at peak exercise. The resting heart rate in the supine position was 70 ± 7 bpm and increased to 113 ± 9 bpm at peak exercise. Although the heart rate was not affected by posture during exercise, it was profoundly influenced by posture at rest and during recovery. Both at rest and during recovery, it was significantly lower in the supine position than in the upright position (Fig. 2A).

Figure 2: Changes in heart rate (HR) (A), oxygen uptake (V̇O₂) (B), and oxygen pulse (V̇O₂/HR) (C) during upright and supine exercise. The value of recovery “0” was measured at the end of exercise. R, rest. *P < 0.05; **P < 0.01 by Student’s paired t-test; ††P < 0.01 by ANOVA.

Oxygen uptake and oxygen pulse.

Figure 2B shows the oxygen uptake during exercise and recovery. Oxygen uptake in the supine position did not differ from that in the upright position and showed a similar response from rest until 10 min after recovery irrespective of posture. In both supine and upright exercise, oxygen pulse (oxygen uptake/heart rate) increased during exercise with increasing exercise intensity and then quickly returned to the preexercise level during recovery (Fig. 2C). The oxygen pulse was significantly higher at rest preceding exercise in the supine position, but there was no significant difference during exercise. During recovery, however, it was higher in the supine position than in the upright position, although the difference was not significant (P = 0.08).

Diastolic filling velocity.

Figure 3 A demonstrates the peak A velocity during exercise and recovery. The peak A velocity did not differ between the upright and supine positions. Although the peak E velocity was significantly higher in the supine position than in the upright position in both the resting (P < 0.01) and recovery states (P < 0.05), there were no significant differences in the peak E velocity during exercise (Fig. 3B). The ratio of E to A peak velocities (E/A) in the upright position was 1.52 ± 0.13 at rest and decreased to 1.34 ± 0.11 at peak exercise (Fig. 3C). In the supine position, the E/A ratio was 1.88 ± 0.29 at rest and decreased to 1.42 ± 0.16 at peak exercise. The ratio was significantly higher in the supine position than in the upright position in both the resting and recovery states (P < 0.01). However, the ratio did not differ between the upright and supine position during exercise. The higher E/A ratio at rest and during recovery in the supine position resulted mainly from the higher E wave.

Figure 3: Changes in peak A velocity (A), peak E velocity (B), and peak E to A ratio (C) during upright and supine exercise. The value of recovery “0” was measured at the end of exercise. R, rest. **P < 0.01 by Student’s paired t-test; †P < 0.05; ††P < 0.01 by ANOVA.

DISCUSSION

In the clinical setting, measurements of the left ventricular filling velocities using Doppler echocardiography have always been made in the supine position. Because postural change must profoundly affect stroke volume through the Frank-Starling mechanism (²⁵), it could also influence the left ventricular filling velocities. These influences on the filling velocities may be more accentuated or even attenuated during exercise as compared with the resting state. However, there had been no direct comparison of the filling velocities between supine and upright exercise. Thus, we sought to investigate the effects of postural change on the left ventricular filling velocities both at rest and during incremental exercise.

In the present study, in both the upright and supine positions, the E/A ratio gradually decreased with exercise intensity and then quickly returned to the preexercise resting value after exercise. The E/A ratio was significantly higher in the supine position at rest and during recovery from exercise, mainly as a result of the higher E wave in the supine position. During exercise, however, the difference in the E/A ratio between the supine and upright positions was negligible.

Peak E velocity was significantly higher in the supine position than in the upright position in both the resting and recovery states, but there was no significant difference during exercise. Thadani et al. (³⁵) reported that left ventricular filling pressure and pulmonary capillary wedge pressure are both higher in the supine position than in the upright position, probably due to increased venous return. Their findings are consistent with the higher peak E wave velocity in the supine position noted in our study, because peak E wave velocity reflects left atrial pressure as well as the left ventricular relaxation (^5,15). It has also been demonstrated by Stoddard et al. (³²) that peak E wave velocity was increased by an increase of preload in normal subjects.

The peak A wave velocity did not significantly differ between the two positions throughout the test. The physical determinants of peak A wave velocity include left atrial contraction (¹⁹), left atrial pressure (¹³), and the degree of afterload (¹). Left ventricular chamber distensibility, which reflects afterload, is known to be well preserved in normal subjects regardless of posture. Therefore, our finding that the peak A wave velocity was not influenced by posture in normal young subjects does not contradict the previous reports (⁴).

The change of oxygen uptake closely reflects the change of cardiac output by Fick’s equation. In the present study, oxygen uptake in the upright position did not differ from that in the supine position throughout the test. Therefore, it can be presumed that the cardiac output was similar between the two positions. Although we did not measure stroke volume, we measured oxygen pulse (oxygen uptake/heart rate) as a representative of stroke volume (³³). Oxygen pulse in the supine position was significantly higher than that in the upright position at rest (P < 0.01) and had a tendency to be higher during recovery. This finding, which suggests a higher stroke volume in the supine position at rest and during recovery, can be attributed to a higher venous return together with the Frank-Starling mechanism (²⁵).

As reported previously, the resting heart rate in the upright position was higher than that in the supine position. The higher heart rate in the upright position appears to be the result of a compensation for the lower stroke volume. Heart rate during recovery after exercise was also higher in the upright position. However, the posture did not influence heart rate during exercise. This may be related to the augmented stroke volume during exercise in the upright position. Stroke volume during upright exercise was probably maintained at a level similar to that of supine exercise. Also, venous return during exercise, which can be presumed to be reflected in the E/A ratio, did not differ between upright and supine exercise. The absence of any postural influence on oxygen pulse and E/A ratio during exercise in our study was probably due to the presence of the muscle pump (^18,23) and the higher catecholamine level during exercise (²⁶).

Although measurement of left ventricular filling velocities is completely noninvasive and clinically useful for evaluating diastolic function, it was found that the filling velocities were profoundly influenced both by the posture and by the exercise intensity. Especially, the E/A ratio was significantly higher in the supine position than in the upright position in both the resting and recovery states. The higher E/A ratio was primarily caused by the higher peak E velocity, reflecting increased left ventricular filling pressure. Although E/A ratio in the present study was within the physiologic range and thus its difference between supine and upright position may not be physiologically significant, the potential influences of posture and exercise intensity should be taken into account when interpreting diastolic function by Doppler echocardiography.

This work was supported in part by a grant-in-aid for Scientific Research from the Ministry of Education, Science and Culture of Japan. We would like to thank Hiromasa Adachi, M.D., and Toshihiko Takamoto, M.D., of Hokushin General Hospital for their valuable contributions. We also appreciate the valuable assistance of Takasuke Imai, M.D., of Tokyo Medical and Dental University.