Mechanisms of pulsus alternans

Pancycle aortic incompetence and double alternans in pediatric heart failure

A 12-year-old boy presented with bicuspid aortic valve, severe aortic regurgitation, and dilated dysfunctional left ventricle in heart failure. He underwent aortic valve replacement with a 23 mm TTK Chitra heart valve prosthesis (tilting disk). He was gradually weaned off milrinone and noradrenaline in the intensive care. Echocardiography showed severe left ventricular dysfunction with an ejection fraction of 24%. The radial pulse was regular and of normal volume but exactly half that of the heart rate. Evaluation of the rhythm and echocardiography revealed an interesting hemodynamic phenomenon with double alternans.

Cardiac Alternans: From Bedside to Bench and Back

Cardiac alternans arises from dynamical instabilities in the electrical and calcium cycling systems of the heart, and often precedes ventricular arrhythmias and sudden cardiac death. In this review, we integrate clinical observations with theory and experiment to paint a holistic portrait of cardiac alternans: the underlying mechanisms, arrhythmic manifestations and electrocardiographic signatures. We first summarize the cellular and tissue mechanisms of alternans that have been demonstrated both theoretically and experimentally, including 3 voltage-driven and 2 calcium-driven alternans mechanisms. Based on experimental and simulation results, we describe their relevance to mechanisms of arrhythmogenesis under different disease conditions, and their link to electrocardiographic characteristics of alternans observed in patients. Our major conclusion is that alternans is not only a predictor, but also a causal mechanism of potentially lethal ventricular and atrial arrhythmias across the full spectrum of arrhythmia mechanisms that culminate in functional reentry, although less important for anatomic reentry and focal arrhythmias.

Occurrence of pulsus alternans during anaesthesia of two dogs and one cat and its treatment

The authors report the occurrence of pulsus alternans, a condition characterised by the alternance of pulses of higher and lower amplitude, in two dogs and one cat under general anaesthesia. The presence of an underlying cardiac disease was confirmed in the cat but not in either dog, which - based on history and clinical findings - had presumably normal cardiovascular function before the anaesthetic. Possible mechanisms, including negative inotropy and haemodynamic and Frank-Starling effects, as well as the role of general anaesthesia as the potential triggering factor, are discussed in this report. Ephedrine resulted in the successful treatment of pulsus alternans, as demonstrated by the return of normal pulse and synchronisation of heart and pulse rates in the cat and in one dog. In the other dog, pulse pattern and frequency returned to normal once the guidewire for central line placement was withdrawn.

Non-Invasive Detection of Mechanical Alternans Utilizing Photoplethysmography

BACKGROUND AND SIGNIFICANCE

Mechanical alternans (MA) is a biomarker associated with mortality and life-threatening arrhythmias in heart failure patients. Despite showing prognostic value, its use is limited by the requirement of measuring the continuous blood pressure (BP), which is costly and impractical.

OBJECTIVE

To develop and test, for the first time, non-invasive MA surrogates based on photoplethysmography (PPG).

METHODS

Continuous BP and PPG were recorded during clinical procedures and tests in 35 patients. MA was induced either by ventricular pacing (Group A, N = 19) or exercise (Group B, N = 16). MA was categorized as sustained or intermittent if MA episodes were observed in at least 20 or between 12 and 20 consecutive beats, respectively. Eight features characterizing the pulse morphology were derived from the PPG, and MA surrogates were evaluated.

RESULTS

Sustained alternans was observed in 9 patients (47%) from Group A, whereas intermittent alternans was observed in 13 patients (68%) from Group A and in 10 patients (63%) from Group B. The PPG-based MA surrogate showing the highest accuracy, V^'_M, was based on the maximum of the first derivative of the PPG pulse. It detected both sustained and intermittent MA with 100% sensitivity and 100% specificity in Group A and intermittent MA with 100% sensitivity and 83% specificity in Group B. Furthermore, the magnitudes of MA and its PPG-based surrogate were linearly correlated (R² = 0.83, p < 0.001).

CONCLUSION

MA can be accurately identified non-invasively through PPG analysis. This may have important clinical implications for risk stratification and remote monitoring.

Pulsus alternans in a child with dilated cardiomyopathy

A previously healthy 21-month-old presented with new-onset dilated cardiomyopathy. Evaluation noted pulsus alternans, with beat-to-beat alternations in aortic pulse wave amplitude, both clinically and on diagnostic studies. Pulsus alternans is an infrequent, complex pathophysiologic sign often associated with severe heart failure. The mechanisms are incompletely understood, but theorised aetiologies include beat-to-beat changes in left ventricular loading conditions, variations in myocardial oxygen supply/demand, and alternations in myocardial contractility. Recognition of pulsus alternans is important as it provides significant clinical information, may suggest suboptimal medical management, and may be the first warning sign of severe cardiac dysfunction.

Biventricular pulsus alternans: An echocardiographic finding in patient with pulmonary embolism

Pulsus alternans is characterized by regular rhythm with beat-to-beat alternation of systolic pressures. Left ventricular alternans is usually found in severe left ventricular dysfunction due to cardiomyopathy, coronary artery disease, systemic hypertension, and aortic stenosis. Right ventricular alternans is usually associated with left ventricular alternans, right ventricular dysfunction, pulmonary embolism, and pulmonary hypertension. Biventricular alternans is rare and associated with severe left ventricular dysfunction and left anterior descending coronary artery disease. The exact mechanism of pulsus alternans has not been clearly delineated, and it has been remained a subject of investigation and conjecture since the nineteenth century. Two fundamental mechanisms have been proposed to explain ventricular alteration. The first, based on the Frank-Starling mechanism, proposes beat-to-beat alteration in end-diastolic volume accounted for the alternating contractile force. The second proposed mechanism which explains the physiology of pulsus alternans involves the abnormal calcium handling by cardiac myocytes. To the best of our knowledge, biventricular alternans in pulmonary embolism has not been previously reported in the medical literature. We present and discuss the mechanisms of pulsus alternans and its clinical implications.

Murmur and Doppler alternans in critical pulmonary stenosis

Pulsus alternans is a well-recognized clinical entity in which alternating strong and weak pulses are detected. It usually is secondary to underlying myocardial failure. Murmur alternans (alternation in murmur intensity) has been described in aortic stenosis and a few right-sided lesions such as pulmonary hypertension and embolism. This report describes a case of murmur alternans in critical pulmonary stenosis that also showed Doppler alternans on echocardiography. The underlying cause was right ventricular systolic dysfunction.

Postextrasystolic contractile decay always contains exponential and alternans components in canine heart

In isolated, blood-perfused canine hearts, postextrasystolic potentiation (PESP) decays monotonically after a noncompensatory pause following a spontaneous extrasystole (ES). The monotonic PESP decay yields myocardial internal Ca(2+) recirculation fraction (RF). We have found that after a compensatory pause (CP), PESP decays in alternans, consisting of an exponential and a sinusoidal decay component. We have proposed that this exponential component also yields RF. In the present study, we examined the reliability of this alternative method by widely changing the ES coupling interval (ESI), CP, and heart rate in the canine excised, cross-circulated left ventricle. We found that all PESP decays consisted of the sum of an exponential and a sinusoidal decay component of variable magnitudes whether a CP existed or not. Their decay constants as well as the calculated RF were independent of the ESI and CP. This confirmed the utility of our alternative RF determination method regardless of the ESI, CP, and heart rate. Direct experimental evidence of Ca(2+) dynamics supportive of this alternative method, however, remains to be obtained.

Postextrasystolic transient contractile alternans in canine hearts

We found that postextrasystolic potentiated contractility after a spontaneous extrasystole most frequently decayed as a transient alternans over several beats in excised, cross-circulated, atrially paced canine hearts. This type of heart preparation; which we have been using consistently in mechanoenergetic studies, had normal coronary blood perfusion pressure as well as flow and mechanoenergetic performance. Spontaneous atrial and ventricular extrasystoles occurred occasionally in every heart. Arrhythmic changes in left ventricular (LV) pressure at a fixed volume reflected corresponding changes in contractility. We analyzed nearly 3,600 cases of postextrasystolic potentiation in 68 hearts; 84% decayed as transient alternans, 6% decayed exponentially, and 10% belonged to neither type. We found that a postextrasystolic compensatory pause always preceded the transient alternans after either an atrial or ventricular extrasystole at any constant atrial pacing rate (85-188 beats/min). The decay was either exponential or nonalternating when the pause did not exist after an atrial extrasystole during occasional pacing failure. Therefore, the compensatory pause after either an atrial or ventricular extrasystole seems essential for the postextrasystolic transient alternans of LV contractility in the type of canine heart preparation we have been using.

Cardiac alternans: diverse mechanisms and clinical manifestations

OBJECTIVES

The purpose of this review is to assemble the widely dispersed information about cardiac alternans and to categorize the types and mechanisms of alternans, their clinical manifestations and possible therapeutic implications.

BACKGROUND

The phenomena of mechanical and electrical alternans have been of continuing interest to both physiologists and clinicians. Recent studies have enhanced this interest because of the reported association of alternans with experimental myocardial ischemia and cardiac arrhythmias.

METHODS

The review formulates concepts based on extensive review of published studies and personal observations.

RESULTS

Cardiac alternans has been subdivided into the following four categories: 1) mechanical, 2) electrical, 3) in association with myocardial ischemia, and 4) in association with cardiac motion. Mechanical alternans can be explained by hemodynamic or inotropic alterations, or both. Mechanical alternans in the ventricular muscle is accompanied by alternans of action potential shape. In the Purkinje fibers, action potential duration alternates without change in shape and is determined by the duration of the preceding diastolic interval. However, in ventricular muscle fiber, alternans can occur in the presence of constant diastolic intervals. T wave alternans reflects changes in action potential duration and is frequently associated with a long QT interval. Electrocardiographic manifestations of conduction alternans occur at many different sites within the conducting system and myocardium. During myocardial ischemia, additional mechanisms of repolarization alternans have been proposed. Alternans occurring in the presence of a large pericardial effusion is attributed to swinging motion of the heart maintaining two-beat periodicity.

CONCLUSIONS

Since its origin as "pulsus alternans" described by Traube in 1872, the definition of alternans has evolved into a term encompassing multiple physiologic and pathologic phenomena that, although united by the term cardiac alternans, diverge widely with respect to etiology, mechanism and clinical significance.

Electromechanical concordance in alternans during sustained tachycardias

Electrical alternans and mechanical alternans are intriguing phenomena that have been reported to occur during tachycardias. Their precise pathophysiologic mechanism and in particular their interrelation have not been well defined. This report documents an unusual electromechanical concordance in alternans during sustained supraventricular tachycardia and ventricular tachycardia in two patients, raising interesting possibilities on the underlying mechanism of the concordant phenomena. The findings highlight the principle of excitation--contraction coupling of the heart.

Stability analysis of one-dimensional dynamical systems applied to an isolated beating heart

In this paper we propose a new model of an isolated beating heart. The model is described by a one-dimensional non-linear discrete dynamical system which depends on several parameters. Applying stability analysis we investigate the dynamic properties of the non-linear system. We find those domains in the parameter space in which the equilibrium point of the system (the fixed point) and the periodic orbits are attractors and in which they are unstable. These domains correspond to a normal and abnormal beating heart, i.e. when the end diastolic volumes are stable time invariant and time variant, respectively. On transition between these domains there is a bifurcation which gives rise to a pair of attracting points of period 2. This case corresponds to the simplest type of period doubling behavior of an abnormal beating heart, called mechanical alternans. Our results provide qualitative and quantitative predictions which can be used for comprehensive experimental design.

Evidence for a link between mechanical and electrical alternans in acutely ischaemic myocardium of anaesthetized dogs

In order to examine the relation between mechanical alternans and associated electrical alternans during acute myocardial ischaemia, we determined the effect of a ventricular premature beat and calcium antagonists on mechanical and electrical alternans during acute coronary occlusion in anaesthetized dogs. Isometric contractions and unipolar electrocardiograms were recorded from ischaemic myocardium. During coronary occlusion, mechanical alternans was accompanied by electrical alternans, which was an alternate change in the ST segment elevation, i.e. the higher ST and the lower ST. Electrical alternans was frequently discordant and in some cases accompanied by discordant mechanical alternans. Both discordant electrical and mechanical alternans became concordant and were potentiated after the ventricular premature beat. In all cases, concordant mechanical alternans was accompanied by concordant electrical alternans and vice versa. In this situation, the higher and the lower ST corresponded to the larger and the smaller contractions respectively. Thus, a fixed correspondence was observed between mechanical and electrical alternans. A fixed correspondence was also observed between mechanical alternans and the variation in the time taken for repolarization of the monophasic action potential. Verapamil and diltiazem inhibited both electrical and mechanical alternans. The present results support the idea that a common mechanism, such as a beat-to-beat cycle of the transmembrane and intracellular movement of calcium ions, may play a role in the mechanisms of electrical and mechanical alternans.

Clinical and hemodynamic characteristics of patients with inducible pulsus alternans

Pulsus alternans can be found in some patients with abnormal left ventricular function and also can develop after spontaneous premature beats. The purposes of this study were to: (1) determine the inducibility of pulsus alternans in a series of patients referred for routine cardiac catheterization and (2) define the clinical and hemodynamic characteristics of those who develop pulsus alternans. In 104 patients referred for right and left heart catheterization, atrial premature beats and rapid atrial pacing were used to try to provoke pulsus alternans. The 29 patients who developed pulsus alternans in response to these maneuvers were older (63 +/- 6 vs 59 +/- 10 years, p less than 0.01) and had a greater incidence of valvular heart disease (45% vs 23%, p less than 0.01) and congestive heart failure (38% vs 17%, p less than 0.05). Aortic stenosis was the most prevalent valve lesion found. Those who developed pulsus alternans in response to pacing were further characterized by higher left ventricular systolic (143 +/- 42 vs 121 +/- 23 mm Hg, p less than 0.02) and end-diastolic pressures (17 +/- 9 vs 13 +/- 6 mm Hg, p less than 0.05), higher pulmonary artery systolic pressure (35 +/- 14 vs 29 +/- 11 mm Hg, p less than 0.04), and lower left ventricular ejection fractions (0.42 +/- 0.13 vs 0.53 +/- 0.14, p less than 0.001). Eight patients (28%) with inducible pulsus alternans had a normal left ventricular ejection fraction (greater than 0.50) and left ventricular end-diastolic pressure (less than 13 mm Hg).(ABSTRACT TRUNCATED AT 250 WORDS)

Pulsus alternans induced by inferior vena caval occlusion in man

To assess the effect of rapid preload reduction on left ventricular performance in nonischemic cardiomyopathy, 11 patients were studied during inferior vena caval (IVC) balloon occlusion. Five developed sustained pulsus alternans. During pulsus alternans, the strong beats demonstrated systolic performance characteristics similar to baseline values, despite a drop in both left ventricular (LV) end-diastolic diameter (66 +/- 13 to 61 +/- 13 mm; p less than 0.05) and LV end-diastolic pressure (21 +/- 8 to 9 +/- 6 mmHg; p less than 0.05). In contrast, the weak beats demonstrated a reduction in peak systolic pressure (130 +/- 36 to 109 +/- 33 mmHg; p less than 0.02), fractional shortening (20% +/- 4% to 17% +/- 9%; p less than 0.05) and peak positive dP/dt (1,006 +/- 224 to 921 +/- 287 mmHg; p less than 0.05). Measures of diastolic performance (peak negative dP/dt, the time constant of LV relaxation, the length of diastasis, and LV end-diastolic stress) were not different between baseline beats and the strong beats; and only LV end-diastolic stress differed when baseline beats were compared to the weak beats. When the strong beats were compared to the weak beats during induced pulsus alternans, significant differences were observed in peak systolic pressure, peak positive dP/dt, and fractional shortening, but no differences in any measured diastolic parameter was observed. A slight difference was noted in the left ventricular end-diastolic diameters, with the weak beat consistently beginning at a slightly smaller diameter (61 +/- 13; mm vs 59 +/- 13; p less than 0.05). In summary, these data are consistent with an augmentation and deletion of intrinsic contractile forces in association with an alternation in preload on a beat-to-beat basis as best describing left ventricular performance during pulsus alternans.

Hemodynamics and myocardial oxygen metabolism of pulsus alternans in patients with dilative cardiomyopathy

The hemodynamic and myocardial energetic changes due to pulsus alternans were investigated by left and right heart catheterization and by oxygen consumption measurements in three patients with dilative cardiomyopathy. In all three patients, pulsus alternans developed after intravenous administration of the phosphodiesterase inhibitor enoximone. Following enoximone (Patients 1/2/3), left ventricular peak systolic pressure was reduced, in the respective patients, from 100/103/115 mmHg (normal beat) to 91/96/94 mmHg (strong beat) and further to 59/80/85 mmHg (weak beat); left ventricular end-diastolic pressure was reduced from 24/23/22 mmHg (normal beat) to 5/10/6 mmHg (strong beat) and further to 3/7/4 mmHg (weak beat). Cardiac output increased by an average of 16%. Heart rate increased by an average of 12%. Stroke work (during pulsus alternans mean between strong and weak beats) did not change (less than 5%) in any of the three patients. Arterial-coronary-sinus oxygen content difference decreased by 5%/13%/22, respectively. Myocardial oxygen consumption per beat decreased in Patient 1 by 8%, in Patient 2 by 8% and remained unchanged in Patient 3. It is concluded that pulsus alternans occurred in consequence of alternating systolic performance. The alternation in systolic performance most probably resulted from a disturbance in excitation-contraction coupling induced by enoximone. The pronounced reduction of left ventricular preload following administration of enoximone may have augmented further the differences between the strong and the weak beat. A disturbance in myocardial oxygen metabolism was ruled out as the cause of pulsus alternans in these patients.

Model of mechanical alternans in the mammalian myocardium

A model is proposed to elucidate the cause and mechanism of mechanical alternans in cardiac muscle in terms of discrete calcium movements. Mechanical alternans, the cause of which lies within the borders of excitation-contraction-coupling (ECC), is analyzed. In this case, the "input" of the ECC system (the action potentials and intervals) is constant while the "output" (contractile force) oscillates between two constant values, indicating that the system has a "memory" with two "internal states". It is proposed that these two "states" are associated with a part of the sarcoplasmic reticulum ("releasable terminal") containing the readily releasable calcium. A mechanism of "calcium-concentration-dependent threshold" is suggested to govern the "release function", i.e. the release of calcium from the "releasable terminal" to the myofilaments. The "release function" is analyzed in both the linear and the non-linear cases and its implication on the initiation of sustained and transient mechanical alternans are described. The dependence of mechanical alternans on a disturbance is also explained. The model response resembles the experimental observations of mechanical alternans in mammalian myocardium in the following manners: abrupt transition from low to high heart rates, slow progressive acceleration of rate, variations in persistence at subthreshold rates, effect of premature and delayed beat following the small and large beats, restitution curves, and transient mechanical alternans initiated by a delayed beat.

Mitral valve alternans

An echogram showing mitral valve alternans was recorded in a patient with advanced cardiomyopathy and pulsus alternans induced by premature ventricular depolarizations. Decreased left ventricular emptying by the weak beat was followed by blunting of the subsequent mitral valve E point, a finding compatible with a significant elevation of left ventricular pressure in early diastole. Both alternating systolic (contractile state) and diastolic (pressure and dimension) variables seemed to participate in the alternans phenomenon in this patient.

Pulsus alternans: its influence on systolic and diastolic function in aortic valve disease

Left ventricular high fidelity pressure measurements and simultaneous biplane cineangiocardiography were performed in 12 patients with severe aortic valve disease (aortic stenosis in 10, aortic insufficiency in 1 and combined valve lesion in 1). Left ventricular contractility was estimated from maximal rate of left ventricular pressure rise (max dP/dt), peak measured velocity of contractile element shortening (Vpm) and mean circumferential fiber shortening velocity. Left ventricular relaxation was assessed in 12 patients from the time constant (T) of the decline in left ventricular pressure; this constant was calculated from a nonlinear regression analysis of pressure and time (method 1) and a linear regression analysis of pressure and negative dP/dt (method 2). Left ventricular diastolic function was evaluated in nine patients from simultaneous diastolic pressure-volume relations during the strong and weak beats. During pulsus alternans, heart rate and left ventricular end-diastolic pressure remained unchanged, whereas peak systolic pressure (220 versus 204 mm Hg, p less than 0.01) and end-systolic pressure (101 versus 95 mm Hg, p less than 0.01) were significantly higher during the strong beat than during the weak beat. Max dP/dt was alternating (2,162 versus 1,964 mm Hg, p less than 0.05), whereas the peak velocity of contractile element shortening remained unchanged (1.21 versus 1.18 ML/s). Systolic shortening of the left ventricular minor axis was significantly (p less than 0.02) greater during the strong (24%) than during the weak (19%) beat, but that of the left ventricular major axis remained essentially unchanged (8 versus 7%).(ABSTRACT TRUNCATED AT 250 WORDS)

Acute massive (pericardial effusive) pulmonary thromboembolism--pulmonary embolectomy revisited

The most recent patient in a 13-year experience with 14 patients suffering massive pulmonary thromboembolism requiring pulmonary thromboembolectomy is the focus of this report. This 40-year-old woman not only survived life-threatening acute hypoxemia and right heart failure, but was also found to have developed a unique transudative 700 cc pericardial effusion. Pulmonary artery pressure was 90/30 (mean 50 mmHg), accompanied by 17 mm right ventricular alternans. Systemic alternans and tamponade physiology were absent. This unusual natural model for acute right heart failure illustrates a novel mechanism for pericardial effusion physiology.

Atrial alternans: experimental echocardiographic and hemodynamic demonstration during programmed pacing

Simultaneous hemodynamic and echocardiographic recordings were used to demonstrate mechanical atrial alternans during programmed atrioventricular (A-V) pacing in five open chest dogs. Each animal was studied in two stages, first with the A-V conduction system intact (phase I) and later after the experimental induction of complete A-V block (phase II). Atrial alternans was demonstrated during rapid atrial stimulation at cycle lengths ranging from 250 to 120 ms. During phase I, rapid atrial pacing resulted in complex combinations of variable A-V conduction disturbances with superimposed atrial and ventricular alternans. During phase II, atrial alternans could be observed during a programmed prolonged pause in ventricular activity. It is anticipated that this method will facilitate recognition of atrial alternans in various clinical situations and shed further light on its possible hemodynamic significance.

Effect of postural changes on pulsus alternans: an echocardiographic study

We describe a patient who had severe left ventricular dysfunction and pulsus alternans following an episode of severe viral myocarditis. Pulsus alternans was mild with the patient in the supine and squatting positions but increased dramatically on standing. Echocardiograms were used to study changes in left ventricular dimension and function with changes in posture. On standing, there was a decrease in the left ventricular diameter and an increase in the absolute and relative shortening fraction and the velocity of shortening on the strong beats, and there was a decrease in end-diastolic dimension and the measurements of ventricular contractility on the weak beats. The marked increase in pulsus alternans during standing may have been induced by increased sympathetic stimulation following the decrease in ventricular loading conditions, while the great beat-to-beat differences in end-diastolic dimension suggest, in addition, the importance of Starling's mechanism.

Electrical and mechanical response in biventricular mechanical alternans

Mechanical alternans of various degrees is produced by rapid heart rates, slower rates in failing hearts and can be brought about by a single extra systole. It has also been shown that the two ventricles may exhibit different degrees of mechanical alternation. The present study was planned to clarify the possible mechanism inducing this latter phenomenon. For this reason myocardial tension was recorded simultaneously from the two ventricles as well as through a miniature strain gage capable of measuring electrogram and myocardial tension of a small area -- just adjacent to a stimulating electrode. The heart was driven at a steady heart rate through one electrode and very late premature beats were applied at various coupling times at another site through an electrode attached to the miniature strain gage. It was found that the degree of mechanical alternans is markedly different at the sites of measurements in either ventricle. These changes could be related to the time interval elapsed between the application of the electrical stimulus and the occurrence of the mechanical response.

The demonstration of alternating contractile state in pulsus alternans

Pulsus alternans was induced in 11 anesthetized, open-chest dogs by rapid atrial pacing, and the left ventricular filling characteristics and length-tension-velocity relationship of alternating beats were compared. The end-diastolic circumferences (cire) of the strong beats were slightly, but significantly, increased over the weak beats (7.3 > 6.9 cm, P < 0.01), confirming that diastolic filling does alternate in pulsus alternans. This alternation in initial fiber length seemed to result from an alternation in the prior end-systolic length, rather than from an alternation in diastolic filling time or compliance. There was also no difference in end-diastolic tension as measured by an isometric strain gauge suggesting no difference in contractile element relaxation before weak and strong beats. THE CONTRACTILE STATE OF THE STRONG BEATS WAS CONSISTENTLY GREATER THAN THAT OF THE WEAK BEATS WHEN CONTRACTILITY WAS DEFINED IN TERMS OF: (a) V(max) (3.13 > 2.53 circ/sec, P < 0.01); and (b) the velocity of circumferential fiber shortening (0.84 > 0.39 circ/sec, P < 0.001) and developed tension (82.5 > 74 g/cm, P < 0.01) at isolength. The length-tension-velocity relationship of the left ventricle also varied between strong and weak beats when: (a) the maximum velocity of contractile element shortening at least common tension (1.68 > 1.28 circ/sec, P < 0.05); and (b) the velocity of circumferential fiber shortening (0.81 > 0.39 circ/sec, P < 0.001) at maximum developed tension were examined. Analysis of the length-tension-velocity characteristics of sequential beats at the onset of alternans in three dogs suggests that an alternation in contractility initiates alternans, with secondary alternations in ventricular filling. Cross-clamping of the aorta in three other dogs essentially eliminated the alternating changes in end-diastolic length and pressure, while the resultant isovolumic contractions continued to demonstrate clear evidence of pulsus alternans in the ventricular systolic pressure, suggesting the persistance of an alternating contractile state. The evidence suggests that an important mechanism in the production and propagation of pulsus alternans, as produced in the intact canine ventricle by rapid pacing, is a beat-to-beat alternation in contractile state with secondary alternations in ventricular filling.