Mechanical Function of the Left Atrium: New Insights Based on Analysis of Pressure–Volume Relations and Doppler Echocardiography

Precise recording of the LA pressure waveform requires the use of a high-fidelity, intravascular pressure transducer. Placement of a micromanometer-tipped catheter into the LA chamber may be conducted directly through the LA body or appendage or indirectly using a proximal pulmonary vein in the experimental laboratory or during open heart surgery. An intraatrial transseptal technique or a retrograde approach through the mitral valve have been used to measure LA pressure in the cardiac catheterization laboratory. 18The LA pressure waveform is composed of three major deflections during normal sinus rhythm (fig. 1). 19After the P wave of atrial depolarization is recorded on the electrocardiogram, the LA contracts, causing an a wave that occurs late in LV diastole. This a wave may be enhanced by preload augmentation (i.e. , Frank-Starling mechanism) or increases in intrinsic LA myocardial contractility. The rate of deceleration of the a wave is an index of LA relaxation. 20With the onset of LV systole, ventricular contraction causes a pressure wave to be transmitted in retrograde fashion by closure of the mitral valve, resulting in a small increase in LA pressure (i.e. , the c wave). This c wave may be more pronounced in the presence of mitral valve prolapse because excessive leaflet motion into the body of the LA occurs during early LV isovolumic contraction. During late LV isovolumic contraction, LV ejection, and the majority of LV isovolumic relaxation, pulmonary venous blood progressively fills the LA and gradually increases LA pressure, resulting in the LA v wave. This v wave may be accentuated during mitral regurgitation or reductions in LA compliance. 21 

Measurement of continuous LA volume has been successfully performed using a variety of techniques. Left atrial volume is often estimated invasively from long and short axis dimensions measured using epicardial orthogonal sonomicrometry assuming prolate ellipsoid LA geometry. 22Left atrial volume determined using this method has been shown to correlate closely with true LA volume measured using water displacement–atrial cast studies, 23but its use is restricted to the experimental laboratory. Noninvasive assessment of continuous LA volume in humans has been conducted using two- or three-dimensional 24echocardiography with automated boundary detection, 18,25,26tissue Doppler echocardiography, 27radionuclide angiography, 28,29cine computed tomography, 30and magnetic resonance imaging. 24,31In contrast to the LA pressure waveform, the LA volume waveform is essentially monophasic. Minimum LA volume occurs immediately after the completion of LA contraction and corresponds closely to the closure of the mitral valve. Maximal LA volume is observed immediately before the mitral valve opens. When combined with high-fidelity measurement of LA pressure, these determinations of continuous LA volume allow assessment of LA function in pressure–volume phase space.

As a result of the multiple deflection morphology of the LA pressure waveform, the steady state LA pressure–volume diagram consists of two loops arranged in a horizontal figure-of-eight pattern that incorporates both the active (A loop) and passive (V loop) components of LA function (fig. 1). 3Beginning at LA end-diastole, the active component of the diagram proceeds in a counterclockwise fashion during atrial systole as blood is ejected from the LA into the LV through the mitral valve. In contrast to the observations in the LV pressure waveform that facilitate the definition of end-diastole, precise identification of LA end-diastole has varied between investigators. An easily detectable nadir in LA pressure may not always occur immediately before the onset of LA contraction as a result of continuous pulmonary venous return during diastasis. Left atrial end-diastolic pressure may be defined as the pressure occurring immediately before atrial contraction that corresponds to the LA end-systolic pressure 32or may be chosen to occur at a fixed time point before peak LA pressure. 33For the sake of this review, we will use the former definition of LA end-diastole pressure and its corresponding volume (EDV). Despite these relatively minor differences in the definition of LA end-diastole, most investigators have reported remarkably similar values of LA stroke volume using pressure–volume relations in vivo . Left atrial end-systole marks the end of atrial contraction and is most often defined by minimal LA volume. Maximal LA elastance (i.e. , the ratio of LA pressure to volume 34) during contraction has also been used to define of LA end-systole in the normal heart analogous to definition of LV end-systole commonly used in LV pressure–volume analysis. 35,36We will define LA end-systolic pressure and volume (ESV) at minimal LA volume in this review.

Identification of LA end-diastole and end-systole on the LA pressure–volume diagram facilitates the calculation of LA stroke volume (i.e. , EDV − ESV) and emptying fraction (i.e. , stroke volume/EDV). Although frequently used to describe LA contractile function, these ejection phase measures of LA pump performance are highly dependent on LA loading conditions and may not be used as strict quantitative indices of LA inotropic state. After the mitral valve closes, LA filling occurs during LV systole and isovolumic relaxation. Left atrial pressure and volume progressively increase as the chamber expands during the reservoir phase, forming the bottom portion of the A loop and the upper portion of the V loop. The area of the A loop represents active LA stroke work 37analogous to LV stroke work defined as the area inscribed by the LV pressure–volume diagram. 38Under normal circumstances, a small amount of blood contained within the LA at end-diastole refluxes into the pulmonary veins during atrial systole. This retrograde pulmonary venous blood flow does not usually appear in the LA pressure–volume diagram because the peristaltic-like configuration of atrial contraction and the unique valve-like anatomy of the pulmonary vein–LA junction minimize atrial regurgitation at normal LA pressures. 39However, increases in the amount of this atrial regurgitant blood flow into the pulmonary veins occur during increases in LA pressure that may falsely elevate LA emptying fraction by reducing minimal LA volume as depicted in the LA pressure–volume diagram.

In contrast to the active part of the LA pressure–volume diagram, the passive component (V loop) proceeds in a clockwise direction over time, indicating that alterations in LA pressure and volume occurring during this period of the cardiac cycle result from external forces acting upon the LA. Total LA reservoir volume is easily determined from the pressure–volume diagram as the difference between maximum and minimum LA volumes obtained by direct examination of the A and V loops, respectively. 20The area of the V loop represents the total passive elastic energy stored by the LA during the reservoir phase 20and is an index of reservoir function. 40Static compliance of the LA may be assessed from the pressure–volume diagram by determining the slope of the line between minimal LA pressure of the A loop and maximal LA pressure in the V loop. 41Decreases in LA compliance are indicated by increases in the slope of this relation. For example, regional myocardial ischemia 42or severe LV dysfunction 41produces a decrease in LA compliance that may be quantified using this method. When LV pressure falls below LA pressure near the end of LV isovolumic relaxation (i.e. , during early LV diastole), the mitral valve opens, and blood that has accumulated in the LA during the reservoir phase flows down a pressure gradient into the LV. Left atrial emptying during this phase of LV diastole results in a rapid decline in LA volume that forms the bottom portion of the V loop and also produces a concomitant rapid increase in LV volume. Additional pulmonary venous return also enters the LA during LV diastasis but does not substantially affect LA volume because this blood flows directly through the open mitral valve. Thus, the LA conduit phase is defined between mitral valve opening and LA end-diastole, and LA conduit volume is calculated as the difference between maximum and end-diastolic volumes (fig. 1). The areas inscribed by the A and V loops and the crossover point between these components of the steady state LA pressure–volume diagram are ultimately determined by the complex interrelation between LA loading conditions, LA and LV inotropic state, the rate and extent of LA relaxation, LA passive elastic properties, and blood flow through the pulmonary circulation. For example, an increase in LA preload or myocardial contractility or a reduction in LA afterload produce a corresponding increase in A-loop area consistent with the performance of greater stroke work. In contrast, a reduction in LA compliance decreases the relative size of the A loop by shifting the crossover point to the left.

Acute alterations in LA loading conditions produced by mechanical or pharmacological techniques may be used to assess intrinsic LA myocardial contractility using LA end-systolic pressure–volume relations (fig. 2) in the isolated 34,43and intact heart. 41,44Changes in LA inotropic state may be assessed by alterations in the slope (E^es) of the LA end-systolic pressure–volume relation 44in a manner analogous to the well-established evaluation of LV contractility using this method. 45For example, LA E^eshas been used to quantify changes in LA inotropic state produced by chronic LV disease 41(fig. 3) or volatile anesthetics 33(fig. 4). Similar to the LV end-systolic pressure–volume relation, 46the LA end-systolic pressure–volume relation has been shown to be a relatively heart rate- and load-independent index of LA contractile state in vivo . 34,44,47The ratio of LA E^esderived from these pressure–volume diagrams and LV elastance (E^LV; determined using the ratio of LA end-systolic pressure and LA stroke volume) also provides a useful index of mechanical matching #between the LA and the LV that quantifies the relation between the contractile state of the LA and forces resisting its ejection (e.g. , LA afterload) 33,41based on a series of elastic chamber models originally described for LV–arterial coupling. 48,49For example, LA–LV coupling is markedly attenuated in the presence of reduced LV compliance in patients with heart failure. 41In addition, this same series of differentially loaded LA pressure–volume diagrams may also be used to determine the dynamic compliance of the LA in response to alterations in load (fig. 2) 22,33,50similar to the methods extensively validated in the LV. 51This technique has been used to describe changes in dynamic LA stiffness produced by surgical maneuvers (e.g. , pericardectomy, 50LA appendage excision 22) and by the administration of vasoactive drugs, including volatile 33and intravenous 52anesthetics.

Analysis of the pulmonary venous blood flow velocity waveform is commonly used with or without concomitant evaluation of transmitral blood flow velocity to determine the severity of LV diastolic dysfunction, 53quantify the degree of mitral regurgitation, 54–56or estimate pulmonary capillary occlusion and mean LA pressures. 57–60The pattern of pulmonary venous blood flow velocity also provides important information about the active and passive mechanical behavior of the LA in the normal and diseased heart (fig. 1). Measurement of the pulmonary venous blood flow velocity may be conducted invasively in the experimental laboratory using Doppler flow probes placed around 61or implanted within 62a pulmonary vein immediately proximal to the LA chamber. However, pulmonary venous blood flow velocity is most often determined noninvasively using transthoracic or transesophageal pulse wave Doppler echocardiography as previous studies 61,62have demonstrated an excellent correlation between this modality and invasively derived techniques. Transesophageal echocardiography has evolved into the preferred noninvasive method for pulmonary venous blood flow velocity analysis 63because the anatomical proximity of the right and left upper pulmonary veins to the esophagus provides optimal imaging windows with minimal ultrasound scatter by intervening tissue. It is important to note that patterns of pulmonary venous blood flow velocity have been shown to be highly dependent of LA loading conditions, LA contractile state, and LV function, 61,64and conclusions about alterations in LA function derived using this methodology require interpretation within the constraints of these potential limitations.

The normal pulmonary venous blood flow velocity waveform is composed of a single small negative deflection that illustrates retrograde flow from the LA into the pulmonary veins (the atrial reversal [AR] wave;figure 1) and two large positive deflections that depict forward flow from the pulmonary veins into the LA chamber. 65Another model of pulmonary venous blood flow analysis using four separate deflections that also incorporates specific flow during diastasis has also been proposed. 66The first positive deflection (S wave) occurs during LV systole and isovolumic relaxation when the mitral valve is closed and displays a biphasic morphology. 61,63,67The magnitude and velocity–time integral of the S wave are indices of LA reservoir function that closely correlate with reservoir volume measured using the steady state LA pressure–volume diagram. For example, reductions in LA compliance observed during primary atrial disease states 21cause declines in S-wave velocity indicative of compromised reservoir function.

The second positive deflection (D wave) of the pulmonary venous blood flow velocity waveform occurs immediately after the opening of the mitral valve. This forward pulmonary venous flow occurs as a result of the rapid drop in LA pressure that accompanies early LV filling and is an index of LA conduit function. 63The peak velocity and velocity–time integral of the D wave are dependent upon the extent of early LV filling. 68Thus, factors that attenuate early LV filling, such as delayed LV relaxation, may reduce D-wave velocity, indicating that LA conduit function has been adversely affected (fig. 5). 69,70The velocity of the D wave is also reduced by the mechanical obstruction to early LV filling observed in patients with severe mitral stenosis, indicating that LA conduit function is dependent on normal mitral valve motion. 71 

Left atrial systolic performance is most often noninvasively evaluated using the A wave of transmitral blood flow velocity (fig. 1). The velocity–time integral of the transmitral A wave correlates with LA stroke volume determined from the LA pressure–volume diagram. The atrial reversal component of the pulmonary venous blood flow waveform is also directly related to LA contraction. Peak AR blood flow velocity and its corresponding velocity–time integral have been shown to correlate closely with mean LA pressure and volume, respectively. 59These data verify that increases in the quantity of atrial regurgitant blood into the pulmonary veins occurs concomitant with elevations in LA pressure associated with increased LA preload, mitral valve disease, or severe LV dysfunction. The AR peak velocity and its velocity–time integral have been combined with peak transmitral A-wave velocity and its respective velocity–time integral to evaluate alterations in LA–LV coupling in patients with elevated LV end-diastolic pressure and LA afterload mismatch resulting from dilated, infiltrative, or hypertrophic obstructive cardiomyopathy. 72,73These variables may also be used to noninvasively estimate LA +dP/dt^maxas an index of LA systolic function. 74 

Several early investigations compared the mechanical properties of isolated and intact atrial and ventricular myocardium and examined the factors that affect LA contractility. These studies demonstrated that the maximum velocity of shortening of LA myocardium was equal to 75or greater than 76,77LV myocardium under similar loading conditions. Left atrial myocardium was also less sensitive to increases in afterload than LV myocardium in vivo . 75Systolic shortening of the LA is primarily dependent on LA preload and inotropic state in the intact heart, 78but LA emptying fraction is reduced and conduit function enhanced when LA diameter exceeds optimal fiber length. 79Alterations in autonomic nervous system activity produce characteristic changes in LA inotropic state that are similar to those observed in the LV. 80For example, increases in LA emptying fraction and LA contribution to LV filling are observed in normal subjects performing a sustained hand grip 81or rapidly standing from a supine position 82in part as a result of activation of the sympathetic nervous system. In contrast, parasympathetic nervous system stimulation reduces LA pump performance, 80although the resulting bradycardia may offset this response by enhancing LA preload and augmenting LA emptying fraction through the Frank-Starling mechanism. 78,83 

In the absence of mitral stenosis, LA afterload is determined primarily by the elastic properties of the LV and the pressure within this chamber. Thus, LA afterload and LA energy expenditure progressively increases as LV diastolic function deteriorates and LV pressure during diastole increases. Up-regulation of the β myosin isoform in atrial myocardium has been observed with increased LA mechanical work 84,85that further augments the Frank-Starling response to LA dilatation. These compensatory actions enhance LA emptying fraction, but increased workload imposed on the LA myocardium by LA afterload mismatch may contribute to the subsequent development of primary LA contractile dysfunction. 47,60,69For example, an initial increase in LA emptying fraction has been observed early in the course of evolving heart failure, 86but LA systolic function eventually becomes severely depressed as LV chamber stiffness and LV end-diastolic pressure continue to increase. Conversely, indirect increases in LV compliance produced by chronic arterial vasodilator therapy act to reduce LA afterload and improve the active contribution of the LA to LV stroke volume in patients with LV pressure-overload hypertrophy resulting from essential hypertension. 87Left atrial remodeling and reduced compliance may also occur in response to LV diastolic dysfunction. These effects contribute to an exaggerated pressure response during small increases in LA volume, restrict pulmonary venous blood flow into the LA during the reservoir phase, and lead to the development of pulmonary edema.

Multiple factors combine to determine LA reservoir and conduit function. Relaxation of the LA chamber and the resultant reduction in LA pressure that occurs immediately after atrial systole facilitates forward flow from the pulmonary veins into the LA during early LV isovolumic contraction. 20,88,89These events produce the early peak of the biphasic S wave of pulmonary venous blood flow velocity observed with Doppler echocardiography. 67Left ventricular systolic function also plays a very important role in determining early LA reservoir function. The cardiac base descends toward the apex during LV systole, acting like a piston to draw additional blood from the pulmonary venous circulation into the LA. 20,29The mitral annulus has been shown to descend approximately 1.3 cm during LV systole in normal subjects, but this annular motion is markedly attenuated in patients with dilated cardiomyopathy, and as a result, early pulmonary venous blood flow may be blunted or absent. 90Transmission of the right ventricular systolic pressure pulse through the pulmonary circulation contributes to the increases in LA pressure and volume observed later during the reservoir phase 64,91and has been shown to be responsible for the second peak of the biphasic pulmonary venous S wave. 67 

Intrinsic LA compliance plays a major role in determining reservoir and conduit function by facilitating venous return from the pulmonary circulation. 5Atrial diseases in which LA compliance is markedly reduced are associated with impaired LA filling. 21,92,93The LA appendage has been shown to be more compliant than the main body of the LA using pressure–volume relations in isolated 94and intact LA preparations. 22,32Temporary clamping 95or surgical excision 22of the LA appendage reduced LA compliance, decreased reservoir function as quantified by declines in the pulmonary venous blood flow velocity S/D ratio, and attenuated the rate of LV rapid filling. These data indicate that the LA appendage plays an important role in LA reservoir function, especially during increases in LA pressure or volume. 94,95The pericardium has also been shown to affect LA distensibility as pericardiectomy increased LA compliance, enhanced early LV filling rate, and augmented conduit to a greater extent than reservoir function in an elegant study using both LA pressure–volume relations and Doppler echocardiographic analyses of transmitral and pulmonary venous blood flow velocities. 50 

Exercise produces characteristic changes in the determinants of LA function in humans. Left atrial myocardial contractility increases and the LA contribution to cardiac output is more pronounced during aerobic exercise 96,97as a result of sympathetic nervous system activation. Left atrial reservoir but not conduit function is augmented during exercise as quantified by pulmonary venous blood flow velocity measurements. 97This increase in LA reservoir function combines with a pronounced reduction in minimum LV pressure resulting from enhanced LV isovolumic relaxation 98to produce a larger LA–LV pressure gradient during early LV filling, 97thereby augmenting LV stroke volume and cardiac output during exercise. In contrast, a compensatory increase in conduit function has been observed concomitant with LA dilatation in well-conditioned athletes at rest compared to normal subjects. 99 

Increases in LA volume and reductions in passive LA emptying have been observed in healthy elderly subjects (aged > 70 yr) studied using a combination of transmitral Doppler and two-dimensional echocardiographic techniques. 100Dilatation of the LA produces a compensatory increase in LA ejection force 101and augments active LA contribution to LV filling. 100,102,103Left atrial storage fraction (defined by the ratio of storage volume of the LA during reservoir phase to total LV stroke volume) increases in elderly patients in association with LA dilatation and is inversely related to LV ejection fraction. 104The LA dilatation observed in elderly patients may contribute to increases in LA wall stress and eventual LA contractile dysfunction. 105 

An isolated reduction in LA compliance that occurs independent of mitral valve disease or LV dysfunction is the pathognomonic finding in patients with stiff LA syndrome. 21,106Left atrial dilatation is a common associated finding in this syndrome, but the increase in LA volume does not coincide with observed decreases in LA compliance. 107Left atrial reservoir function is severely compromised as a result of the noncompliant LA. Cardiac catheterization typically reveals a large LA pressure v wave without evidence of mitral regurgitation or a significant mitral valve gradient. 106Patients with this disorder invariably develop pulmonary hypertension, pulmonary edema, and right ventricular failure because LA filling is profoundly impaired, initially during exercise but later at rest as well. 21A time-varying load model of the pulmonary vasculature supported these clinical observations and indicated that isolated reductions in LA compliance cause increases in pulmonary and LA pressures that are similar to those observed in stiff LA syndrome. 108Severe fibrosis and calcification of the LA are characteristic autopsy findings in these patients. 21 

Loss of LA contraction with the onset of atrial fibrillation is commonly associated with a reduction in cardiac output. An early study by Mitchell and Shapiro 11demonstrated that several compensatory mechanisms are recruited to maintain cardiac output at rest or during mild to moderate cardiac stress (e.g. , exercise) in the presence of atrial fibrillation, but declines in cardiovascular performance occur with the loss of atrial systole during more profound stress or concomitant LV dysfunction. A reduction in LA compliance and an increase in the LA pressure peak v wave have been observed with onset of atrial fibrillation. 18,109The increase in LA pressure enhances the LA–LV pressure gradient during early LV filling to maintain stroke volume in the absence of atrial booster pump function. 110Administration of dobutamine reduces LA chamber stiffness and LA size assessed with LA pressure–area relations in humans with atrial fibrillation (fig. 6), presumably by indirectly improving LA afterload. 18Left atrial reservoir function and early LV filling are maintained by increased LA compliance during infusion of dobutamine despite the reduction in LA preload, and LV stroke volume increases as a result of direct positive inotropic effects on LV myocardium. 18The S wave of pulmonary venous blood flow velocity may be attenuated in patients with atrial fibrillation because LA relaxation after active contraction does not occur. 111Nevertheless, the S wave and its corresponding velocity–time integral provide effective indices of LA filling, and LA reservoir function has been shown to be preserved to a greater extent in patients with isolated atrial fibrillation than in those with dilated cardiomyopathy using this technique. 111 

Atrial fibrillation most often results from sustained increases in LA afterload that cause enlargement of the LA chamber. 112Conversely, progressive LA dilatation also occurs in patients with atrial fibrillation independent of alterations in LV function or geometry. 113As a result, the presence of atrial fibrillation may establish a positive feedback loop that precipitates further LA enlargement, reduces the likelihood that chemical or electrical cardioversion will be successful, and increases the probability of thrombus formation. Atrial fibrillation increases LA myocardial oxygen consumption and coronary blood flow and reduces the peak reactive hyperemic response. 114Coronary vasodilator reserve is not completely exhausted during atrial fibrillation, but limitations in coronary flow reserve may also contribute to development of LA ischemia, fibrosis, and further perpetuation of the arrhythmia. 115Atrial fibrillation and other atrial tachyarrhythmias 116also produce a characteristic biatrial myopathy 117that is very similar to experimental tachycardia-induced cardiomyopathy. 118The presence of this dysfunctional atrial myocardium also contributes to continuation of the atrial fibrillation. 118Thus, preservation of normal sinus rhythm and atrial contraction may eliminate the detrimental effects of LA dilatation, avert the potentially adverse reductions in atrial perfusion, and prevent the development of atrial cardiomyopathy that sustain atrial fibrillation once it has been established. 115,117,119Advanced age has been identified as the major risk factor for the development of postoperative atrial fibrillation that is associated with increased morbidity and prolonged hospitalization. 120 

Atrial stunning occurs after defibrillation from or spontaneous conversion to sinus rhythm after brief or prolonged periods of atrial fibrillation that reduce the active contribution of the LA to LV filling and increases thromboembolic risk. 121,122The degree of contractile dysfunction observed during atrial stunning after cardioversion is inversely related to LA chamber size. 123Reduced LA emptying fraction has been observed after 30 min of atrial fibrillation in dogs. 124Treatment with verapamil attenuated but the calcium channel agonist Bay K 8644 exacerbated the development of atrial stunning, suggesting that this process is mediated in part by intracellular calcium overload. 125As few as several minutes of atrial fibrillation may be sufficient to produce stunned atrial myocardium after cardioversion in humans. 126Verapamil also improved the recovery of LA emptying fraction in this setting. 126However, inhibition of Na⁺–H⁺exchange more effectively reduced the severity of LA stunning in a canine rapid pacing model of atrial fibrillation than nifedipine. 127These data are similar to those observed in stunned ventricular myocardium 128and implicate a role for intracellular acid–base balance and its indirect action on the Na⁺–Ca²⁺exchanger in the pathophysiology of atrial stunning.

Left atrial stunning was demonstrated after spontaneous conversion of atrial fibrillation of only 60 min duration to sinus rhythm in normal canine hearts. 129Interestingly, LA appendage stunning was more prolonged than contractile dysfunction of the LA body, 129suggesting potential mechanisms by which overall LA emptying fraction remains depressed 130and thrombosis may occur in the appendage after cardioversion. 129The LA appendage displays a characteristic pattern of emptying that may be assessed by Doppler echocardiographic measurement of peak outflow blood flow velocity. 122,131Thrombus formation has been shown to be associated with appendage dilatation and contractile dysfunction in patients during sinus rhythm. 132Thus, it is not surprising that LA appendage dysfunction before cardioversion predisposes to thrombosis. 122Left atrial appendage stunning also occurs in patients with atrial flutter, but the risk of thromboembolic events after cardioversion appears to be less in these patients as compared to those with atrial fibrillation, presumably because LA appendage systolic function is maintained to a greater extent. 133 

Electrical cardioversion of atrial fibrillation produces LA pump dysfunction that is more severe and persists for a greater duration than either spontaneous conversion to or pharmacologically induced restoration of normal sinus rhythm. 134For example, cardioversion using amiodarone produces relatively rapid restoration of LA emptying fraction in the vast majority of patients with new-onset atrial fibrillation. 135In contrast, endocardial defibrillation or external cardioversion may produce LA chamber and appendage stunning in these patients. 136–138 

Internal atrial defibrillation in particular results in depressed LA emptying fraction and may cause accumulation of spontaneous echocardiographic contrast (an indicator of blood stasis and a risk factor for thromboembolism) or the development of thrombosis after cardioversion, 136but the severity of stunned atrial myocardium appears to be independent of the electrical energy used for cardioversion in this setting. 137Atrial stunning and transient spontaneous echocardiographic contrast have also been observed after a 15-min episode of atrial fibrillation followed by internal defibrillation in patients with documented cardiac disease. 138Atrial stunning persisted but spontaneous echo contrast resolved rapidly after defibrillation in this study, suggesting that thromboembolic risk may be remain relatively low after restoration of sinus rhythm despite a continued reduction in LA emptying fraction. 138In contrast to atrial stunning observed after defibrillation following a short episode of atrial fibrillation, cardioversion-induced restoration of sinus rhythm in patients with chronic atrial fibrillation is associated with a gradual increase in LA emptying fraction and cardiac output over 4 weeks in the majority of patients. However, cardiac output may decrease initially in some patients. This initial depression of cardiac output may persist for up to a week after cardioversion and contributes to an increased incidence of pulmonary edema and thromboembolic complications. 139Another study demonstrated that LA emptying fraction and reservoir function gradually improve over 3 months after cardioversion of chronic atrial fibrillation. 140 

Reductions in LA emptying fraction occur in patients with idiopathic dilated cardiomyopathy concomitant with LA dilatation consistent with the presence of a primary atrial myopathy. 141These findings contrast with those observed in patients with pressure-overload hypertrophy or ischemic cardiomyopathy in which declines in LA emptying fraction occur primarily as a consequence of increases in LA afterload. 141,142Histologic evidence of atrial fibrosis is more apparent in patients with dilated cardiomyopathy as compared to those with remote myocardial infarction, suggesting that a primary atrial disease process occurs in dilated cardiomyopathy that cannot be attributed solely to LA mechanical overload. 143Exercise capacity is directly related to LA emptying fraction and inversely related to LA volume in dilated cardiomyopathy, observations that emphasize the critical importance of LA pump performance to functional capacity in patients with this disease. 144Marked reductions in LA systolic function and kinetic energy transfer to the LV have also been observed in patients with AL (amyloid light chain) amyloidosis (formerly known as primary amyloidosis) that occur as a consequence of amyloid infiltration into atrial myocardium 92,145and are associated with a grave prognosis. 146However, LA emptying fraction remains relatively normal before amyloid infiltration becomes echocardiographically apparent. 145Declines in LA compliance also occur in both dilated and infiltrative cardiomyopathy that attenuate LA reservoir function, increase pulmonary arterial pressures, and contribute to the development of right ventricular failure.

Experimental models of atrial myopathy provide additional insight into the mechanical consequences of LA contractile dysfunction that mimic many of the features observed in dilated and infiltrative cardiomyopathy. Rapid atrial pacing (400 beats/min) for 1 week in dogs produces an atrial myopathy characterized by impaired global and regional LA systolic shortening with relative preservation of LV function. 147An increase in the transmitral E-to-A ratio and a decrease in the pulmonary venous S-to-D ratio were observed consistent with reduced LA emptying fraction and increased conduit function, respectively. 147Rapid atrial pacing of longer duration (6 weeks) also produces decreases in LA compliance, LA systolic dysfunction, impaired reservoir function, and enhanced conduit function as assessed with LA pressure–volume relations. 148Interestingly, LA failure induced by rapid atrial pacing has little or no effect on cardiac output and right ventricular function if LV function remains normal because increases in conduit function offset reductions in emptying fraction and reservoir capability. 148,149However, augmented conduit function is unable to compensate for impaired LA systolic performance and reservoir function in the presence of concomitant LV diastolic dysfunction. 149This experimental finding may be especially important in dilated and infiltrative cardiomyopathy during which profound abnormalities in LV systolic and diastolic function are observed.

Substantial increases in LA chamber stiffness and reductions in reservoir function that may influence cardiac output have been reported in patients with hypertrophic obstructive cardiomyopathy. 150These declines in LA compliance are proportionally greater in hypertrophic obstructive cardiomyopathy as compared to other forms of pressure-overload LV hypertrophy. 82Declines in LA emptying fraction may also occur earlier in the natural history of hypertrophic obstructive cardiomyopathy. 151Increases in LA afterload produced by hypertrophied LV myocardium and elevations in LV end-diastolic pressure contribute to a reduction in LA emptying fraction and increase retrograde pulmonary venous blood flow during atrial systole. 152Mitral regurgitation during middle and late LV systole also markedly attenuates LA reservoir function by dramatically increasing LA pressure. Patients with hypertrophic obstructive cardiomyopathy demonstrate abnormal echocardiographic indices of LA relaxation and filling. 93These findings support the contention that hypertrophic obstructive cardiomyopathy is a disease that directly affects both atrial and ventricular myocardium 153regardless of the distribution of LV hypertrophy or the severity of LV outflow tract obstruction. 93Nonsurgical septal reduction 154in patients with hypertrophic obstructive cardiomyopathy reduces LA size and improves LA ejection force and kinetic energy expenditure in conjunction with a decline in the LV outflow tract pressure gradient, resolution of mitral insufficiency, and improved LA diastolic function. 155,156These findings are associated with concomitant increases in passive LV filling and exercise capacity. 156 

Passive emptying of the LA is impaired in patients after heart transplantation because of alterations in LV diastolic function in the donor organ. 157As a consequence, LA preload is greater in these patients, and LA stroke volume may be maintained or even augmented by activation of the Frank-Starling mechanism. This effect plays an important role in preserving LV stroke volume despite reductions in intrinsic LA myocardial contractility. 157,158The contractile elements of the donor heart dominate overall LA booster pump function after heart transplantation. Nevertheless, overall LA emptying fraction may be reduced in the transplanted as compared to the normal heart because some dysfunctional LA myocardium remains intact in the recipient. 159Left atrial emptying fraction may also be depressed after heart transplantation as a result of atrial contractile asynchrony because recipient atrial remnants are electrically isolated and contract independent of donor atrial and ventricular myocardium. 160Heart transplantation using selective bicaval and pulmonary venous anastomoses is associated with relative preservation of active and passive LA function as compared to conventional biatrial techniques. 161 

Acute myocardial ischemia or infarction resulting from brief or prolonged occlusion of the left anterior descending coronary artery (LAD) produces LA dilation, enhances LA preload, and increases LA emptying fraction 162by the Frank-Starling effect 37,40that serves to maintain LV stroke volume despite the reduction in LV systolic function. 163These compensatory alterations in LA size and emptying fraction are often manifested by electrocardiographic evidence of LA stress during and after the acute ischemic event. 164Left atrial pressure–area relations derived using a micromanometer and echocardiographic automated boundary detection in patients with isolated LAD stenoses indicate that LV supply or demand ischemia produced by balloon occlusion or rapid pacing, respectively, is associated with enhanced LA stroke work (i.e. , A-loop area) and reservoir function (i.e. , V-loop area) concomitant with increases in LA preload (fig. 7). 42These findings confirmed the well-established role of augmented atrial booster pump function for the maintenance of cardiovascular performance in patients with acute myocardial infarction. 12In contrast, patients with left circumflex coronary artery (LCCA) stenoses of similar severity failed to display enhanced LA emptying fraction but instead demonstrated increases in LA static compliance and conduit function during supply or demand ischemia. 42These observations were attributed to the presence of LA ischemia because coronary arterial blood supply to the LA is derived from branches of the LCCA. 165,166Thus, LA systolic compensation for LV ischemia is adversely affected by the presence of simultaneous LA ischemia. Pressure–volume analysis of adaptation to increases in mechanical load associated with remote myocardial infarction and ventricular hypertrophy also indicated that augmented LA emptying fraction contributes to the preservation of LV filling, but adverse reductions in static LA compliance (fig. 8) and reservoir function were also observed that may limit further compensatory responses, 41especially during exercise.

Left atrial afterload mismatch and impaired LA–LV coupling have been observed in dogs during acute myocardial infarction produced by prolonged LAD occlusion despite simultaneous increases in LA emptying fraction. 167This afterload mismatch may be attributed to elevations in LV end-diastolic pressure resulting from LV diastolic dysfunction. Administration of dobutamine reversed these detrimental effects by further enhancing LA emptying fraction and indirectly reducing LV chamber stiffness via  declines in arterial load. 167Thus, administration of positive inotropic drugs or arterial vasodilators may facilitate more efficient transfer of LA stroke volume to the LV in the presence of ischemic injury. The importance of efficient mechanical matching between the LA and LV after myocardial infarction is further emphasized by the observations that LV dilatation and increased LV end-diastolic pressure also may lead to progressive reductions in LA stroke volume index in patients with remote myocardial infarction. 168Such an inverse correlation between LA stroke volume and LV end-diastolic pressure has also been described in patients with symptomatic coronary artery disease using Doppler echocardiography. 169Left atrial systolic dysfunction associated with afterload mismatch contributes to the development of pulmonary edema and right ventricular failure 170and has been shown to be closely related to secretion of atrial natriuretic peptide as a compensatory response to perceived volume overload. 171Nevertheless, LA emptying fraction may be relatively preserved even in patients with severe ischemic cardiomyopathy, 142in contrast to the findings in those with idiopathic dilated 141or hypertrophic obstructive cardiomyopathy. 151Interestingly, the extent of LA emptying fraction at rest has been shown to predict LV diastolic filling and cardiac output during exercise in patients with recent myocardial infarction. 172These latter findings emphasize the critical role of enhanced LA emptying fraction and relatively normal LA–LV coupling in determining the functional capacity of these patients.

Left ventricular pressure-overload hypertrophy caused by pathologic conditions, such as essential hypertension and aortic stenosis, delays LV isovolumic relaxation, impairs early LV filling, increases LV chamber stiffness, and elevates LV filling pressures. 173These factors combine to increase LA afterload and produce compensatory LA dilatation. 174In the absence of other demonstrable causes, this enlargement of the LA in patients with hypertension appears to be most highly correlated with elevated nocturnal arterial blood pressure 175and may represent an early clinical sign of hypertensive heart disease detected using two-dimensional echocardiography before any electrocardiographic changes become evident. 176Increases in LA reservoir and reductions in conduit function have been observed in patients with long-standing hypertension that occur as a consequence of LV diastolic dysfunction and elevations in LA afterload. 87,174,177Impaired LA–LV coupling resulting from this afterload mismatch has also been quantified in patients with pressure-overload hypertrophy. 72The increase in LA preload associated with LA dilatation contributes to enhanced LA emptying fraction by activation of the Frank-Starling mechanism that is partially responsible for maintenance of LV stroke volume in hypertensive patients. 174Sympathetic nervous system stimulation also appears to play a role in augmented LA inotropic state in patients with essential hypertension, in contrast to the findings in patients with remote myocardial infarction. 40However, chronic increases in active LA workload and energy expenditure may produce LA hypertrophy, reduce LA compliance, compromise reservoir function, and contribute to an eventual reduction in LA pump performance. 87Such a reduction in LA systolic function may eventually contribute to the development of heart failure in patients with pressure-overload hypertrophy. Nevertheless, LA emptying fraction appears to be relatively well-preserved in the vast majority of patients with essential hypertension, in contrast to those with idiopathic dilated 141or hypertrophic obstructive cardiomyopathy 151in whom evidence of atrial myopathy has been demonstrated. Left atrial dilatation may contribute to the development of atrial arrhythmias in patients with hypertension and may be associated with paroxysmal atrial fibrillation and subsequent atrial stunning that significantly impair LV filling. 178Antihypertensive therapy with a diuretic or an angiotensin-converting enzyme inhibitor normalizes alterations in indices of active and passive LA function concomitant with regression of LV hypertrophy. 87Nifedipine has also been shown to reverse hypertension-induced alterations in LA function concomitant with improvements in LV diastolic function in humans. 179 

The contribution of atrial systole to LV filling and performance in patients with severe aortic stenosis is well-known 180as the loss of LA pump function with the onset of atrial fibrillation is often poorly tolerated in these patients. 181Left atrial dilatation in aortic stenosis is directly related to LV mass. 182Frank-Starling–induced increases in LA emptying fraction serve to counterbalance the depressed reservoir function that occurs as a result of decreases in LV compliance. Thus, LA dilation and augmented LA systolic function are important compensatory mechanisms that serve to maintain LV stroke volume and cardiac output in these patients. 182 

The mitral valve does not contribute substantially to resistance of blood flow from the LA to the LV under normal circumstances. However, restricted motion of the mitral apparatus becomes the predominant factor affecting LA afterload in mitral stenosis. Left atrial pressure and volume increase in direct proportion to the severity of the stenosis. Despite the increase in LA preload, the contribution of atrial systole to total LV filling in patients with mitral stenosis is reduced during sinus rhythm in comparison to normal subjects because LA contractile force cannot overcome the mechanical obstruction. 180Intrinsic myocardial contractility of the LA is depressed in long-standing mitral stenosis as a result of chronic elevations in workload and wall stress, 183and the contribution of the LA appendage to overall LA emptying fraction is also reduced. 184As a result, loss of LA contractile function with the onset of atrial fibrillation may be less responsible for hemodynamic decompensation in patients with mitral stenosis than a rapid ventricular response and limited LV diastolic filling time. 25,185 

Left atrial compliance is an important determinant of LA pressure in patients with mitral stenosis during normal sinus rhythm in addition to the pressure gradient across the mitral valve. 186Declines in LA compliance have been shown to correlate with increases in LA pressure and progressive narrowing of the mitral valve orifice. 187These reductions of LA compliance and elevations in LA pressure are associated with compromised reservoir function. Left atrial pressure–area relations have been recorded in patients with mitral stenosis before and after retrograde balloon valvuloplasty that provide important insights into the mechanical consequences of this disease. 188Significant increases in A-loop area were observed after valvuloplasty in the presence of sinus rhythm consistent with enhanced emptying fraction and stroke work (fig. 9). In contrast, V-loop area increased after the procedure in patients with atrial fibrillation, indicating that reservoir function had been improved. Increases in LA static compliance also occurred concomitant with reductions in pressure and volume in both groups. 188These data confirm that mitral stenosis produces profound alterations in LA function that may be acutely reversed in large part with valve repair.

The effects of chronic mitral regurgitation on LA function have been examined using LA pressure–dimension relations. 189Left atrial size and mass increase during chronic mitral regurgitation, and the LA contribution to LV filling is augmented in sinus rhythm as a result of activation of the Frank-Starling mechanism. The LA also becomes more compliant, and reservoir function may be enhanced. Thus, enlargement of the LA is an important compensatory mechanism in chronic mitral regurgitation by attenuating increases in LA pressure while simultaneously maintaining adequate LV filling volume. 189However, LA volume overload and pronounced dilatation of the chamber may eventually lead to reductions in LA emptying fraction because optimal myocardial fiber length is exceeded. This concept is emphasized by the observation that LA size alone predicts outcome after mitral valve replacement in patients with symptomatic chronic mitral regurgitation because LA size reflects the severity and duration of the disease process. 190The v-wave magnitude of the LA pressure waveform has been shown to be inversely related to LA compliance, and increases in the amplitude of the v wave indicate increasing severity of acute mitral regurgitation and decreased LA compliance during incremental balloon commissurotomy. 191 

Regurgitant blood flow during LV systole attenuates, abolishes, or reverses early LA expansion and forward flow from the pulmonary veins dependent upon the degree of the regurgitation. 25This process contributes to the development of pulmonary hypertension and right ventricular dysfunction. The pulmonary venous blood flow velocity pattern assessed with Doppler echocardiography has been shown to be a very sensitive index of the severity of mitral regurgitation that is highly correlated with angiographic grade of valve disease. 54,55Acute mitral regurgitation of increasing severity is initially associated with augmented LA emptying fraction as a result of LA dilatation. However, the LA contribution to LV filling rapidly declines as the regurgitant fraction increases because LA volume overload occurs. Treatment with sodium nitroprusside improves LA emptying fraction during severe acute mitral regurgitation by decreasing excessive LA stretch, restoring LA geometry, and reestablishing more optimal Frank-Starling relations. 192Enhanced LA emptying fraction occurring as a consequence of vasoactive drugs in acute mitral regurgitation most likely results from reductions in LV afterload, decreases in LV volume overload, and improvements in mitral valve geometry and competence. 193 

Left atrial pressure–volume analysis of LA adaptation to evolving LV failure produced by rapid ventricular pacing has been examined in a canine model of dilated cardiomyopathy. 47The development of heart failure over 3 weeks of pacing was associated with progressive increases in LA volume, stroke volume, and A-loop area (LA stroke work). Myocardial contractility evaluated with end-systolic pressure–volume relations was unchanged, but LA mean circumferential fiber shortening was reduced in a time-dependent manner. An up-regulation of the β myosin heavy chain was also observed concomitant with decreased velocity of LA contraction and increased mechanical work. 47These latter findings indicate that compensatory increases in LA emptying fraction initially occur during developing LV failure. Temporal improvements in LV systolic function were observed after cessation of rapid ventricular pacing in this canine model, but LA systolic ejection rate was persistently depressed as a result of continued LV diastolic dysfunction, LA hypertrophy, and alterations in myosin heavy chain isoforms. 194Intrinsic LA dysfunction quantified using a variety of invasive and noninvasive techniques eventually occurs in heart failure because persistent increases in LA afterload and energy expenditure resulting from reduced LV compliance and elevated LV diastolic wall stress are present. 41,86,183,195–197Interestingly, contractile function of the LA appendage may be an accurate predictor of LV end-diastolic pressure in patients with heart failure. 198Left atrial emptying fraction is inversely and LA maximal volume is directly related to plasma renin activity, aldosterone concentration, and atrial natriuretic peptide concentration in patients with heart failure resulting from idiopathic dilated cardiomyopathy, 199suggesting that hormonal compensatory responses are correlated closely with LA function under these conditions. Primary LA systolic failure may be observed even during normal sinus rhythm in the presence of severe, long-standing LV dysfunction because of complete exhaustion of contractile reserve. 200Nevertheless, treatment of heart failure may be associated with improvements in LA emptying fraction as LV stiffness declines, LA afterload mismatch is reduced, and LA–LV coupling is normalized. 41,86For example, afterload reduction acutely enhances LA emptying fraction in the failing heart. 86This finding supports the hypothesis that declines in LA performance occur principally as a result of LA afterload mismatch during LV failure and not as a consequence of primary LA pathology.

Profound alterations in the passive mechanical properties of the LA are also observed during the development of heart failure. Left atrial reservoir function was augmented and conduit function was reduced in patients with a minor derangement in LV diastolic function as indicated by an impaired relaxation transmitral blood flow velocity pattern. 69In contrast, patients with a restrictive LV filling pattern indicative of severe LV diastolic dysfunction and elevated LV diastolic pressure demonstrated a predominance of LA conduit function and a marked reduction in reservoir function concomitant with contractile dysfunction (fig. 8). 69These findings indicate that evolving heart failure is associated with the progressive conversion of the LA from a storage and contractile chamber to a simple conduit. This concept was dramatically emphasized by the report of a patient with amyloidosis and end-stage heart failure in whom complete LA akinesis was observed throughout the cardiac cycle despite the continued presence of an electrocardiographically demonstrable normal sinus rhythm. 92The ratio of conduit to active LA emptying volume has also been shown to be greater in patients with normal as compared to pseudonormal transmitral LV filling patterns. 201These data suggest that analysis of active and passive LA function provides an alternative means of distinguishing between normal and pseudonormal filling patterns that may be used instead of standard analysis of pulmonary venous blood flow. 53 

Left atrial pressure–area relations also demonstrate that LA compliance is reduced in patients with congestive heart failure and normal sinus rhythm or atrial fibrillation. 18,41,202Improvements in LA distensibility have been observed with the administration of positive inotropic drugs 18or arterial vasodilators 203in congestive heart failure and, conversely, abnormal LA compliance is further exacerbated by β-adrenoceptor antagonists or additional LA preload in this setting. 202Effects of dobutamine and sodium nitroprusside on LA function were examined in patients with severe congestive heart failure using Doppler echocardiographic evaluation of transmitral and pulmonary venous blood flow velocity. 204Dobutamine increased LA reservoir and conduit volumes but did not substantially affect pump function, alter pulmonary arterial pressures, or influence the restrictive pattern of LV filling that was observed under baseline conditions in these patients. In contrast, sodium nitroprusside did not alter reservoir or conduit volume but did enhance LA contractile performance and improve the pattern of LV filling from a restrictive to a normal morphology. These findings suggest that an arterial vasodilator may acutely provide more consistent improvements in LA and LV function than a β¹-adrenoceptor agonist by reducing LV afterload, improving LV compliance, decreasing ventricular interaction, 205restoring LA preload reserve, and enhancing LA emptying fraction. 204 

The negative inotropic effects of halothane and methoxyflurane were initially described by Paradise et al.  206–208in rat atrial myocardium in vitro . Volatile anesthetics also depress the contractile function of atrial myocardium obtained from guinea pigs, 209rabbits, 210and humans. 211–213These actions have been attributed to reductions in transsarcolemmal calcium (Ca²⁺) influx through voltage-dependent Ca²⁺channels and decreases in Ca²⁺availability from the sarcoplasmic reticulum, 210mechanisms that are very similar to those responsible for anesthetic-induced depression of LV myocardium. 214The negative inotropic effects of volatile agents in the intact LA were recently quantified using pressure–volume analysis. 33Desflurane, sevoflurane, and isoflurane reduced LA contractility (i.e. , E^es) by approximately 50% at an end-tidal concentration of 1.2 minimum alveolar concentration (MAC;fig. 4). The magnitude of this effect in LA myocardium was similar to the degree of LV contractile depression produced by these agents as quantified with LV end-systolic pressure–volume relations. 215Desflurane, sevoflurane, and isoflurane also impaired LA and LV relaxation to similar degrees. These data indicate that volatile anesthetics produce equivalent alterations in contractility and relaxation in LA compared to LV myocardium. 33The magnitude of reductions in LA inotropic and lusitropic state produced by the volatile anesthetics was also similar in the intact LA, supporting the results obtained in isolated human atrial myocardium. 213 

Desflurane, sevoflurane, and isoflurane altered LA passive mechanical behavior. 33Left atrial reservoir function (V-loop area and reservoir volume) was maintained during the administration of anesthetic concentrations of less than 1.0 MAC (fig. 10). This preservation of reservoir function contributed to the relative maintenance of LV stroke volume 215by compensating for decreases in LV filling associated with a reduced contribution of LA contraction. The volatile anesthetics also reduced dynamic LA chamber stiffness, an action that most likely contributed to the preservation of reservoir function because the delays in LA relaxation and declines in LV systolic function that also occurred would be expected to decrease reservoir function. 20However, LA reservoir function was reduced during administration of higher concentrations of the volatile anesthetics because further impairment of LA relaxation and LV contractility occurred. Decreases in the ratio of LA stroke work to total pressure–volume diagram area and the increases in the ratio of LA conduit to total reservoir volume (fig. 11) were also produced by desflurane, sevoflurane, and isoflurane. These data indicated that the LA contribution to LV filling becomes less active and more passive during the administration of the volatile agents.

Desflurane, sevoflurane, and isoflurane decreased the ratio of LA to LV elastance (E^es/E^LV), consistent with impaired mechanical matching between these chambers. Volatile anesthetics have been shown to produce LV diastolic dysfunction by delaying LV isovolumic relaxation and impairing early LV filling in association with direct negative inotropic effects. 216Thus, the attenuation of transfer of kinetic energy from the LA to the LV probably resulted from the combination of LA contractile depression and LV systolic and diastolic dysfunction. Volatile anesthetic-induced abnormalities in LA–LV matching were greater than analogous impairment of LV–arterial coupling evaluated using a similar series of elastic chamber models in a previous investigation 215because these agents produced beneficial alterations in the determinants of LV afterload 217,218that partially compensate for simultaneous depression of LV myocardial contractility.

Propofol depresses the contractile function of isolated atrial myocardium obtained from guinea pigs 219and humans 220at concentrations higher than those typically achieved during intravenous infusions in a clinical setting. These findings are similar to those observed in normal ventricular myocardium in vitro , 221,222,in situ , 223and in vivo . 224,225The negative inotropic actions of propofol in ventricular myocardium have been attributed to inhibition of transsarcolemmal calcium (Ca²⁺) current 226,227and L-type Ca²⁺channel function, 228and it is likely that similar mechanisms are responsible for depression of contractility in atrial myocardium. The effects of propofol on LA myocardial contractility in vivo  were recently quantified using pressure–volume analysis. 52The magnitude of the negative inotropic effect of several doses of propofol in LA myocardium was nearly identical to the degree of LV contractile depression. 225Dose-related declines in E^es/E^LVwere observed during administration of propofol consistent with impaired mechanical matching between these elastic chambers. In contrast to the findings with volatile anesthetics, previous investigations 229,230have demonstrated that propofol does not affect LV relaxation and compliance. Thus, the impairment of LA–LV coupling observed during the administration of propofol most likely resulted from the depression of LA contractile function and not because of LV diastolic dysfunction.

Propofol also affected LA passive filling and emptying properties. 52Increases in V-loop area occurred during administration of larger doses of propofol, and total LA reservoir volume was unchanged. These findings suggested that LA reservoir function is maintained during propofol anesthesia. This preservation of reservoir function may partially compensate for reductions in the active contribution of LA contraction to LV filling and serves to maintain stroke volume. 225Left atrial chamber stiffness decreased during administration of propofol despite modest increases in LA pressure, suggesting that LA compliance is improved by this intravenous agent. The preservation of reservoir function that occurred during administration of propofol was probably related to these decreases in LA chamber stiffness, because decreases in LV systolic function were observed that would be expected to reduce reservoir function. 20A delay in LA relaxation has also been shown to contribute to a reduction in reservoir function, 20but an LA relaxation constant was unchanged during administration of propofol. These latter data support previous observations indicating that this drug does not alter LV relaxation. 229,230 

Insights obtained from the analysis of LA pressure–volume relations and Doppler echocardiography have substantially advanced our understanding of LA function in the normal and diseased heart. The active and passive mechanical actions of the LA play critical roles in determining overall cardiovascular performance by unloading the pulmonary venous circulation and by facilitating LV filling. Compensatory LA enlargement and enhanced LA emptying fraction produce an increase in the active LA contribution to LV filling in a variety of pathologic conditions that serve to maintain stroke volume and cardiac output. However, LA dilatation may eventually adversely affect LA emptying fraction if optimal Frank-Starling relations are exceeded or atrial tachyarrhythmias occur. In addition, declines in LA compliance adversely affect LA reservoir function, impede pulmonary venous blood flow into the LA, and impair LV filling. Left atrial failure may occur as a consequence of primary atrial disease or chronic elevations in LA afterload and contribute to the development of clinical signs and symptoms of congestive heart failure. Pharmacological management of heart failure not only enhances LV function, but also improves the interaction between the LA and LV. Volatile and intravenous anesthetics have recently been shown to profoundly affect LA function in the normal heart. How these agents influence LA mechanical behavior in the presence of LV dysfunction remains an important goal of future research.

The authors thank David A. Schwabe, B.S.E.E. (Senior Research Scientist, Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, Wisconsin), for his assistance.