| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Cardiovascular Biophysics Laboratory and 2Section of Applied Physiology, Washington University School of Medicine, St. Louis, Missouri 63110
Submitted 22 January 2004 ; accepted in final form 14 June 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 peak), is predictive of mortality and the necessity for cardiac transplantation in patients with chronic heart failure (HF). However, the role of left ventricular (LV) diastolic function at rest, reflected by chamber stiffness assessed echocardiographically, as a determinant of exercise tolerance is unknown. Increased LV chamber stiffness and limitation of O2 peak are known correlates of HF. Yet, the relationship between chamber stiffness and O2 peak in subjects with HF has not been fully determined. Forty-one patients with HF New York Heart Association [(NYHA) class 2.4 ± 0.8, mean ± SD] had echocardiographic studies and O2 peak measurements. Transmitral Doppler E waves were analyzed using a previously validated method to determine k, the LV chamber stiffness parameter. Multiple linear regression analysis of O2 peak variance indicated that LV chamber stiffness k (r2 = 0.55) and NYHA classification (r2 = 0.43) were its best independent predictors and when taken together account for 59% of the variability in O2 peak. We conclude that diastolic function at rest, as manifested by chamber stiffness, is a major determinant of maximal exercise capacity in HF. diastole; echocardiography; oxygen consumption
400,000 new cases of HF diagnosed each year, 4.7 million Americans suffer from this debilitating disease (1). It has been established that patients with HF have impaired exercise tolerance that is predictive of mortality and the need for cardiac transplantation (10). Characteristically, the failing heart demonstrates an inability to maintain an adequate cardiac output, first during effort and later also at rest. Cardiac decompensation ultimately leads to impaired exercise capacity, fatigue, and exertional dyspnea. Patients with HF thus may have a decreased left ventricular (LV) ejection fraction (LVEF), large end-diastolic volume, and impaired contractile reserve (33, 34). Ventricular dysfunction as quantitated by a decreased LVEF is the main manifestation of HF that ultimately results in circulatory failure. Advanced LV dysfunction is associated with a variety of neurohumoral, peripheral circulatory, skeletal muscle, and respiratory adaptations that determine the syndrome's overall clinical presentation and prognosis (33, 34). The known mechanisms that lead to HF are highly complex, involving chamber enlargement and remodeling and decreased LV contractile reserve (10). The factor common to all types of HF regardless of etiology is increased LV end-diastolic pressure (LVEDP), which compromises LV filling, diminishes stroke volume and cardiac output, and is one of the earliest clinical signs of HF when, due to elevation of the alveolar-arterial (A-a) gradient, it manifests as dyspnea.
The best marker of exercise capacity is oxygen consumption (O2) at peak exercise (O2 peak) (22). Changes in O2 peak are mediated by cardiac output and/or oxygen extraction from exercising skeletal muscles. LV diastolic function can modulate cardiac output by determining filling via chamber stiffness and LVEDP. Therefore, it is plausible that impaired LV diastolic function at rest, manifesting as increased chamber stiffness, resulting in increased LVEDP and associated increase in A-a gradient, can limit aerobic power in HF. Kitzman et al. (14) demonstrated that reduced O2 peak in patients with HF and normal ejection fraction was primarily due to abnormalities of diastolic function that limited LV filling and consequently reduced the stroke volume response to exercise. Moreover, LV systolic performance has not been shown to be highly correlated with exercise tolerance in patients with HF (21, 24). However, the relationship between resting diastolic function expressed in terms of chamber stiffness determined noninvasively and exercise capacity in patients with HF is unclear. As motivation for the present study, we hypothesized that LV chamber stiffness could be the diastolic abnormality that is a significant determinant of aerobic capacity in patients with HF and systolic dysfunction. By using a validated kinematic model to analyze the filling process, applied to different New York Heart Association (NYHA) class patients, we determined O2 peak and diastolic function at rest as reflected by LV chamber stiffness (16).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 peak determination (11). All patients received standard therapy for HF. Patients had taken their medications, including angiotensin-converting enzyme inhibitors and 
-blockers, before the O2 peak evaluation. Ninety-eight percent of patients were on -blockers, 88% on an angiotensin-converting enzyme inhibitor, and 66% on digoxin. Because our study sought the physiological relationship between diastolic function and O2 peak in HF patients as encountered in clinical practice, we made no attempts to control or modify patients' medications in this cohort.
Echocardiography.
A standard two-dimensional Doppler echocardiography examination was performed immediately before the O2 peak test. Specifically, the examination was performed with the patient supine or in the left lateral position. Transmitral Doppler flow velocity images (E waves) were acquired in the four-chamber view according to standard American Society of Echocardiography criteria using a commercial echocardiographic imaging system (model SONOS 5500, Hewlett-Packard, Andover, MA) equipped with a 1- to 3-MHz transducer (29).
Determination of peak aerobic power.
A maximal treadmill exercise test was performed using the following protocol: after a warm-up during which the patients walked on a treadmill at 0% grade for 3–4 min at speeds ranging from 0.5 to 1.2 miles/h, speed was kept constant, and the grade was increased 1 or 2% every 1 or 2 min until the patients could no longer continue to exercise either because of exhaustion or cardiac symptoms or because they showed ECG changes or a drop in systolic blood pressure (
10 mmHg) that justified cessation of exercise. O2 peak was measured continuously with open-circuit spirometry, as previously described (15). Maximal O2 (O2 max) was defined using the following criteria: 1) an attainment of plateau of O2 despite increasing exercise intensity, 2) a respiratory exchange ratio 1.1, and 3) measured O2 lower than predicted for the work rate (15). Because some patients did not meet the technical criteria for O2 max, the highest O2 attained was designated as O2 peak.
Data analysis. Chamber stiffness was determined using a kinematic model of filling that has been previously validated via simultaneous cardiac catheterization and echocardiography in a large subject sample (12, 16, 17, 19). Briefly, the model treats E-wave generation as being initiated by mechanical suction, i.e., the release of elastic strain stored during systole (12, 16, 17). Three mathematically independent model parameters (xo, c, k) are determined from the clinical Doppler E wave by model-based image processing (MBIP). The parameters account for the amplitude, width, and rate of decay of the E-wave contour, respectively (Fig. 1) (16). The model parameter k reflects average chamber stiffness (ratio of change in pressure to change in volume) as determined during simultaneous catheterization and echocardiography (17, 19).
|
O2 peak, our list of dependent parameters consisted of NYHA class, ratio of peak early transmitral flow velocity to peak atrial flow velocity (E/A ratio), peak E-wave velocity, peak A-wave velocity, maximal heart rate, LVEF, age, acceleration time, body mass index, diastolic duration, deceleration time, E-wave duration, resting heart rate, resting systolic blood pressure, and the MBIP-determined parameters xo, c, and k. For each of the 17 selected dependent parameters, univariate regression was initially performed. A stepwise multiple regression analysis was subsequently performed on the most relevant five variables that correlated with O2 peak. At each step in the analysis, variables were entered until no further improvement in the value of the generalized r2 was seen. In accordance with convention that the ratio of subjects (n = 41) to the nine independent predictors be in the 6–10 range, a final set of five independent predictors was used (see Table 2). The variables with a high degree of multicolinearity, quantified as a variance inflation factor of >2.0, were excluded (31). Analyses were performed using a commercially available statistical package (SPSS for Windows version 10.0, SPSS). Data are expressed as means ± SD. P values of <0.05 were considered statistically significant.
|
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 peak, and resting ejection fraction and a higher resting heart rate than other NYHA classes. Moreover, these individuals differed also in E-wave characteristics: increased Epeak (peak of the E wave), decreased acceleration time, and decreased deceleration time. These findings are in agreement with other studies and are consistent with the "constrictive-restrictive" (tall-narrow) E-wave pattern encountered in significant HF (2, 18, 25, 32). The trend in the magnitude of k in our subjects reflects the fact that as NYHA class increases so does LV chamber stiffness (class I, k = 167.8; class II, k = 216.2; class III, k = 253.8; and class IV, k = 307.9).
|
O2 peak, all clinical, hemodynamic, exercise, and Doppler echocardiographic data were proposed for multiple regression model inclusion. Our statistical approach is commonly used in the literature to analyze similar physiological relationships. Specifically, it is similar to the analysis used by Lapu-Bula et al. (18) in their study assessing the relation of exercise capacity to LV systolic function and diastolic filling in idiopathic or ischemic dilated cardiomyopathy. Univariate predictors of O2 peak are reported in Table 2. Age, seated heart rate, diastolic duration, xo, c, gender, body mass index, acceleration time, and deceleration time were not predictors of O2 peak. The parameter k and NYHA classification were the two best predictors of O2 peak, and they were both inversely correlated with O2 peak. Although complete characterization of the Doppler E wave according to the model includes determination of xo (reflecting E-wave amplitude), c (reflecting chamber viscosity and/or relaxation), and k (reflecting chamber stiffness), only k accounted for the majority of variance.
As expected, O2 peak is strongly correlated with NYHA classification in all subjects (P < 0.001) (Fig. 2), but it did not predict NYHA classification as well as did k (P < 0.001) (Fig. 3).
|
|
O2 peak. In this model, k (multiple r2 = 0.55, P < 0.0001) and NYHA classification (multiple r2 = 0.59, P < 0.0001) together predicted 59% of the variability in O2 peak with NYHA classification contributing 4%. |
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 peak. We found strong correlations between k, O2 peak, and NYHA classification. We also found that k appears to be a better predictor of O2 peak than NYHA classification or LVEF most likely because the former is a better surrogate of exercise capacity than the latter two. Patients with HF have a limited ability to perform physical exercise, and this study sheds light on the role of diastolic function as part of that limitation. The recent study by Zile et al. (37) supports our observation that subjects with HF have abnormal diastolic function. More specifically, they concluded that abnormal relaxation and increased passive stiffness are key determinants of diastolic HF (37). Although our patients had systolic dysfunction in addition, the presence of diastolic dysfunction as indicated by the noninvasively derived k accentuates the pathophysiological HF scenario. As a consequence of the demand for increased cardiac output in response to exercise, heart rate increase, diastolic duration shortens, and the associated increase in chamber stiffness (k) mandates development of elevated LVEDP, resulting in dyspnea and compromised oxygenation due to an increase in the A-a gradient with concomitant limitation of cardiac output. This physiological scenario may account in part for the observation that LV chamber stiffness determined by the same method was the strongest independent predictor of 1-yr mortality in elderly subjects hospitalized with HF (27). Furthermore, Doppler tissue imaging has shown that the Doppler E-wave maximum-to-mitral annular velocity maximum ratio and exercise capacity demonstrate a strong negative correlation. The Doppler E-wave maximum-to-mitral annular velocity maximum ratio has not only been observed to be highly correlated with LVEDP or pulmonary capillary wedge pressure but also has been shown to be a causal determinant of LVEDP as a consequence of the "constant-volume" property of the two-chambered heart (20, 25, 30).
Exercise intolerance in HF patients is a poorly understood multifactorial pathological entity and is mediated by numerous mechanisms encompassing central limitations (i.e., cardiac limitations) peripheral limitations that are predominantly muscular and metabolic in origin, or a combination of both (18). Apart from age, gender, and habitual physical activity that affect O2 peak, the relative contribution of central and peripheral factors to the modulation in O2 peak continues to be investigated (6). Central components include cardiac output, heart rate, mean arterial pressure, stroke volume, LV end-diastolic and end-systolic volumes, LVEF, and the Doppler echo-determined E/A ratio. Miyashita et al. (24) recently reported that LV diastolic function, but not systolic function during exercise, determined exercise capacity in patients with HF, indicating that adequate LV filling appears to be more important than LV emptying to provide an adequate stroke volume needed during exercise. In a review by Little et al. (21), it was noted that cardiac output during exercise depended on the capacity of the LV to increase its ability to fill without an abnormal increase in left atrial pressure. However, in patients with HF, the ability to enhance cardiac output and LV filling without a concomitant increase in LV filling pressure during exercise was lost. Because central hemodynamic factors by themselves may not be sufficient to explain the limitation of exercise in patients with HF, interest in determining the extent to which peripheral factors also play a role continues to evolve.
Peripheral factors that may contribute to exercise intolerance include abnormalities in skeletal muscle blood flow and early anaerobic metabolism. Our finding contrasts with previous studies that found selected peripheral components rather than central hemodynamic parameters to be better predictors of O2 peak in HF (5, 8, 9, 13, 23). For example, Cicoira et al. (5) demonstrated that skeletal muscle is an important determinant of exercise capacity independent of central hemodynamics in noncachectic patients with HF. Leg muscle cross-sectional area has also been shown to be significantly correlated with and serve as an independent predictor of O2 peak in male patients with HF (13). Furthermore, exercise capacity is limited in part by the fixed delivery of blood to the periphery, especially in subjects with small muscle mass where local metabolic factors dominate (23). Thus there are data to support the view that both central and peripheral responses are involved in limiting exercise capacity in patients with HF.
The exertional fatigue exhibited in patients with HF has been attributed to underperfusion of skeletal muscle because patients with HF on average demonstrate reduced cardiac output and leg blood flow responses to exercise (23, 27–29, 31–36). Patients with HF and exertional fatigue were, however, shown to be limited by skeletal muscle abnormalities rather than by skeletal muscle underperfusion (35). Moreover, other more recent studies indicate that early skeletal muscle anaerobic metabolism is the primary factor limiting exercise performance in patients with HF (3). Hence, uncertainty remains in characterizing exercise intolerance pathophysiologically. Even though cardiac pump function and peripheral abnormalities may exacerbate the other, the question of which is the primary cause remains ambiguous. In an attempt to find an answer, Florea et al. (7) found that, in patients with severe HF, aerobic performance was primarily determined by peripheral hemodynamic factors, whereas, in patients with mild to moderate HF, exercise tolerance was predominantly mediated by central hemodynamic factors. Thus the novel finding in this study is that increased chamber stiffness determined at rest is the measure of diastolic dysfunction that predicts exercise capacity in subjects with systolic dysfunction. Our results support the perspective that central factors may be dominant in determining values of O2 peak.
Limitations.
In light of previous work relating peripheral factors to exercise limitation, we did not specifically measure peripheral factors. Accordingly, our findings that LV chamber stiffness explained the major portion of the variance in O2 peak as a measure of exercise tolerance in HF subjects with decreased LVEF should be viewed as complementary to previous studies demonstrating the role of peripheral factors in exercise limitation in HF patients.
Although we found a strong linear relationship between O2 peak and k in subjects from all four NYHA classes, several additional limitations remain.
We examined a modest number of patients who were on heterogeneous therapy at different stages in their disease process, and the vast majority of patients were treated with angiotensin-converting enzyme inhibitors and -blockers. The medications, the underlying cause of HF, local hormonal changes, and the variability among the NYHA classes among the patients may have an effect on the observed relationship.
We did not measure LV diastolic stiffness during exercise but determined it at rest. It is known that increased LV diastolic stiffness as manifested via increased LVEDP is disadvantageous during exercise and that it represents an unfavorable hemodynamic condition for exercise by sharply compromising LV filling. Although we observed a high correlation between O2 peak and k, it is possible that a slightly different correlation would be obtained if k were measured during exercise.
Because of respiratory and heart rate variation, there exists a slight beat-to-beat variability of E-wave amplitudes, whereas the variation in E-wave duration is minimal. It is known that the amplitude (peak) of the E wave is preload dependent (3, 4). However, the MBIP method used to estimate k accounts for preload (i.e., E-wave amplitude) effects through the (mathematically) independent model parameter xo, which was found to be an insignificant predictor of O2 peak (12, 16, 17).
The mean age of the patients in this study is somewhat young for a HF population, although it reflects the population in the HF clinic at our institution. The MBIP method of chamber stiffness determination used in this study is a well-established (noncommercial) research tool for which an approximate conventional surrogate in the form of transmitral flow-based deceleration time has also been shown to correlate with poor outcome in patients with HF (2). The strength of MBIP resides in its ability to causally characterize the determinants of the entire Doppler contour. In contrast, current methods utilize only one or two points from the entire contour and cannot distinguish between differently shaped E-wave contours that may have the same peak and deceleration time.
Although we have not yet completed determination of normal gender- and age-related values for k, the trend our study reveals is meaningful, in that HF subjects with stiffer ventricles tend to have worse O2 peak.
Conclusion.
This study shows that HF patients with increased ventricular chamber stiffness determined at rest have worse exercise tolerance. The chamber stiffness determined at rest accounts for a majority of the statistical variance in O2 peak and suggests that it is a major factor contributing to limitation of functional capacity of patients with HF. Moreover, because this index can be obtained noninvasively, it may be a helpful adjunct to exercise testing to assess exercise capacity in clinical practice. Further studies in larger groups of patients with specific etiologies of HF should be undertaken.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
O2 max to training in 60–71 yr olds. J Appl Physiol 71: 2004–2011, 1991.70 years of age with congestive heart failure. Am J Cardiol 84: 82–86, 1999.[CrossRef][Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  M. Noel, J. Jobin, A. Marcoux, P. Poirier, G. R. Dagenais, and P. Bogaty Can prolonged exercise-induced myocardial ischaemia be innocuous? Eur. Heart J., July 1, 2007; 28(13): 1559 - 1565. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Shmuylovich and S. J. Kovacs E-wave deceleration time may not provide an accurate determination of LV chamber stiffness if LV relaxation/viscoelasticity is unknown Am J Physiol Heart Circ Physiol, June 1, 2007; 292(6): H2712 - H2720. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Chung, A. Strunc, R. Oliver, and S. J. Kovacs Diastolic ventricular-vascular stiffness and relaxation relation: elucidation of coupling via pressure phase plane-derived indexes Am J Physiol Heart Circ Physiol, November 1, 2006; 291(5): H2415 - H2423. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Kovacs, J. Toth, J. Tarjan, and A. Koller Correlation of flow mediated dilation with inflammatory markers in patients with impaired cardiac function. Beneficial effects of inhibition of ACE Eur J Heart Fail, August 1, 2006; 8(5): 451 - 459. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Riordan and S. J. Kovacs Quantitation of mitral annular oscillations and longitudinal "ringing" of the left ventricle: a new window into longitudinal diastolic function J Appl Physiol, January 1, 2006; 100(1): 112 - 119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-F. Tse, C.-W. Siu, K. L.F. Lee, K. Fan, H.-W. Chan, M.-O. Tang, V. Tsang, S. W.L. Lee, and C.-P. Lau The Incremental Benefit of Rate-Adaptive Pacing on Exercise Performance During Cardiac Resynchronization Therapy J. Am. Coll. Cardiol., December 20, 2005; 46(12): 2292 - 2297. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Rovner, N. L. Greenberg, J. D. Thomas, and M. J. Garcia Relationship of diastolic intraventricular pressure gradients and aerobic capacity in patients with diastolic heart failure Am J Physiol Heart Circ Physiol, November 1, 2005; 289(5): H2081 - H2088. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
, angular velocity; t, time.
