Biomedical subjects
Publications and source records attributed to W J Stewart.
Explore the source record for details and available documents.
The minimally invasive procedure is a new surgical technique that uses a small sternal incision. Because of limited surgical exposure, removal of intracavitary air and visual assessment of cardiac function are not possible. We studied the utility of intraoperative transesophageal echocardiography (IOE) before and after cardiopulmonary bypass in 112 patients (mean age 53.1 +/- 15.2 years, 74 males) who underwent minimally invasive valvular surgery. Surgical procedures included 52 isolated mitral valve procedures (49 repairs, 3 prostheses), 58 isolated aortic valve procedures (16 repairs, 26 prostheses, 16 homografts), and 2 combined aortic and mitral valve repairs. Prepump IOE was useful to confirm valve dysfunction and assist determination of arterial cannulation site. Postpump IOE identified intracardiac air in all patients, which was defined as extensive in 58 (52%) cases. Postoperatively, new left ventricular dysfunction was noted in 22 (20%) patients, more often in the group with extensive air by IOE (17 [30%] of 58 patients) compared with those without extensive air (5 [10%] of 54 patients, P =.01). Second pump runs were required in 7 (6%) of 112 patients: 3 cases of residual aortic regurgitation, 1 case of residual mitral regurgitation, and 3 cases with new ventricular dysfunction. No deaths occurred. We conclude that IOE is essential in minimally invasive valvular surgery because it detects problems that require immediate remedy. IOE allows real-time assessment of ventricular filling, ventricular and valvular function, and intracardiac air.
Explore the source record for details and available documents.
The frequent use of diagnostic coronary arteriography and its importance in evaluating results of intervention in clinical trials emphasize the necessity of continued assessment of procedural risk. Several studies have described such risks, but they have often included a diverse group of patients with varying levels of clinical stability. Furthermore, this risk has not been well established in a population of patients with saphenous vein bypass grafts. There is need to define the risk of coronary arteriography in a group of patients who are both clinically similar and stable, and to evaluate the influence of improved technology and increased operator experience on the risk of the procedure. The National Heart, Lung, and Blood Institute-funded Post Coronary Artery Bypass Graft Trial offered the opportunity to evaluate the risk of elective diagnostic coronary arteriography in clinically stable patients studied at two points in time: pre-enrollment and 4-5 years after study entry. In this group of 2,635 angiograms from clinically stable patients over 5 years there were no deaths and the risk of myocardial infarction was 0.08%, while 0.7% had clinically important complications. Non-elective, urgent studies (311 angiograms) on unstable patients were more likely to include angioplasty and were associated with a risk of death of 0.6% and myocardial infarction of 1.3%. Complications did not vary with age or gender. Vascular trauma was more likely to occur using the brachial than the femoral artery entry sites. These results indicate that elective angiography on stable patients can be accomplished with a very low risk of mortality (0% in this study) or serious cardiovascular complication. This supports the safety and usefulness of angiography for clinical intervention trials.
The widespread use and popularity of intraoperative echocardiography (IOE) has resulted from advances in cardiac surgery, reparative procedures for valvular heart disease and, most specifically, mitral valve repair. IOE has grown exponentially and is becoming an integral part of the planning and evaluation of many types of surgical procedures such that it is now considered standard of care especially for the perioperative management of patients undergoing mitral and aortic valve repair. This article discusses the application of intraoperative echocardiography and focus specifically on valvular heart disease as this represents the most widely accepted indication for the procedure in current clinical practice.
OBJECTIVES: To develop and validate an automated noninvasive method to quantify mitral regurgitation. BACKGROUND: Automated cardiac output measurement (ACM), which integrates digital color Doppler velocities in space and in time, has been validated for the left ventricular (LV) outflow tract but has not been tested for the LV inflow tract or to assess mitral regurgitation (MR). METHODS: First, to validate ACM against a gold standard (ultrasonic flow meter), 8 dogs were studied at 40 different stages of cardiac output (CO). Second, to compare ACM to the LV outflow (ACMa) and inflow (ACMm) tracts, 50 normal volunteers without MR or aortic regurgitation (44+/-5 years, 31 male) were studied. Third, to compare ACM with the standard pulsed Doppler-two-dimensional echocardiographic (PD-2D) method for quantification of MR, 51 patients (61+/-14 years, 30 male) with MR were studied. RESULTS: In the canine studies, CO by ACM (1.32+/-0.3 liter/min, y) and flow meter (1.35+/-0.3 liter/min, x) showed good correlation (r=0.95, y=0.89x+0.11) and agreement (deltaCO(y-x)=0.03+/-0.08 [mean+/-SD] liter/min). In the normal subjects, CO measured by ACMm agreed with CO by ACMa (r=0.90, p < 0.0001, deltaCO=-0.09+/-0.42 liter/min), PD (r=0.87, p < 0.0001, deltaCO=0.12+/-0.49 liter/min) and 2D (r=0.84, p < 0.0001, deltaCO=-0.16+/-0.48 liter/min). In the patients, mitral regurgitant volume (MRV) by ACMm-ACMa agreed with PD-2D (r= 0.88, y=0.88x+6.6, p < 0.0001, deltaMRV=2.68+/-9.7 ml). CONCLUSIONS: We determined that ACM is a feasible new method for quantifying LV outflow and inflow volume to measure MRV and that ACM automatically performs calculations that are equivalent to more time-consuming Doppler and 2D measurements. Additionally, ACM should improve MR quantification in routine clinical practice.
BACKGROUND: The impact of echocardiographic-guided treatment on outcome after tricuspid valve (TV) surgery is not well defined. OBJECTIVES: The purpose of this study was to determine clinical and echocardiographic factors associated with adverse outcomes after TV surgery and determine the role of intraoperative echo (IOE) in facilitating successful outcomes after TV surgery. METHODS: Four hundred and one patients (279 females, mean age 60 years) underwent TV surgery and other concomitant cardiac surgery at a single institution and were followed clinically and by echocardiography during a 10-year period. RESULTS: Decreased survival after TV surgery was associated with: preoperative increased New York Heart Association (NYHA) functional classification (relative risk [RR]=2.02), increased left ventricular dysfunction by echocardiography (RR=1.28), and use of a TV replacement strategy (RR=2.92). Decreased event-free survival after TV surgery was associated with concomitant coronary artery bypass grafting (RR=2.97). Late echocardiographic failure (3 to 4+ tricuspid valve regurgitation [TR]) after TV surgery was associated with increased severity of TR on preoperative echocardiogram (odds ratio [OR]=1.91). Decreased late echocardiographic failure after TV surgery was associated with the use of a TV annuloplasty ring with a repair strategy (OR=0.40). The surgical plan was altered at the time of surgery to insure a successful outcome in 32 (10%) of 335 patients based on IOE findings. CONCLUSIONS: Adverse outcomes after TV surgery can be predicted by several preoperative clinical and echocardiographic variables. IOE is useful in improving immediate, but not late, outcomes after TV surgery.
Transmitral Doppler flow patterns of patients with cardiac amyloidosis evolve from an early impaired relaxation to an advanced restrictive pattern. This reflects increasing severity of diastolic dysfunction and hence left ventricular filling pressures. The duration of the pulmonary venous atrial reversal flow was recently shown to exceed that of the mitral inflow A wave in patients with left ventricular end-diastolic pressure greater than 15 mm Hg. The objective of this study was to assess the utility of this index as a measure of the severity of cardiac amyloidosis. Comprehensive transthoracic 2-dimensional and pulsed-wave Doppler echocardiograms of the pulmonary venous and transmitral flows were made of 23 patients (10 women) with biopsy-proven diagnosis of primary systemic amyloidosis and of 49 subjects as age-matched normal controls. The amyloidosis group was divided into non-restrictive and restrictive subgroups on the basis of the patients' transmitral inflow deceleration time (>150 and < or =150 ms, respectively). The durations of the pulmonary venous atrial reversal and mitral inflow A wave were measured, and the differences between the flow durations were compared with the control and published data in the nonrestrictive and restrictive groups. The mean duration of the pulmonary venous atrial reversal was significantly longer in the amyloid than the control group (P < .01). The mean duration of the mitral inflow A wave was significantly shorter in the restrictive group than both the nonrestrictive and the control groups (P < .05). The duration of the pulmonary venous atrial reversal exceeded that of the mitral inflow A wave in all patients with cardiac amyloidosis. The difference in duration between pulmonary venous atrial reversal and mitral inflow A wave was significantly greater in the amyloidosis group compared with the normal group, and this index varied significantly within the amyloid group between the abnormal relaxation and the restrictive groups. The difference in the duration between the pulmonary venous atrial reversal and the mitral inflow A wave is a reliable index of diastolic function and can be used to assess the severity of cardiac amyloidosis.
This study assesses the influence of right ventricular (RV) dilation on the progression of left ventricular (LV) dysfunction and survival in patients with idiopathic dilated cardiomyopathy (IDC). Using transthoracic echocardiography, we studied 100 patients with IDC aged 20 to 80 years (mean 55 +/- 14); 67% were men. In the apical 4-chamber view, diastolic LV and RV chamber area measurements classified patients into 2 groups: group RV enlargement+ (RV area/LV area > 0.5) included 54 patients; group RV enlargement- (no RV enlargement) had RV area/LV area < or = 0.5. Echocardiographic studies were repeated in all patients after a mean of 33 +/- 16 months. At the time of the initial study, the 2 groups did not differ in age, gender, incidence of atrial fibrillation and diabetes, left ventricular mass, and LV ejection fraction, but the RV enlargement+ group had more severe tricuspid regurgitation and less LV enlargement. After 47 +/- 22 months (range 12 to 96), patients in group RV enlargement+ had lower LV ejection fraction (29% vs 34%, p = 0.006) than patients with initial RV enlargement-. At clinical follow-up, mortality was higher (43%) in patients with initial RV enlargement+ than the RV enlargement- patients (15%), p = 0.002. For survivors, the mitral deceleration time averaged 157 +/- 36 ms; for nonsurvivors or patients who required transplant, the mitral deceleration time averaged 97 +/- 12 ms (p < 0.0001). With use of a multivariate Cox model adjusting for LV ejection fraction, LV size, and age, the relative risk ratio of mortality from initial RV enlargement+ was 4.4 (95% confidence limits 1.7 to 11.1) (p = 0.002). Thus, patients with significant RV dilation had nearly triple the mortality over 4 years and more rapidly deteriorating LV function than patients with less initial RV dilation. In IDC, RV enlargement is a strong marker for adverse prognosis that may represent a different morphologic subset.
To investigate time-related trends in the use of preoperative invasive hemodynamics in patients with pure valvular stenosis, the preoperative evaluations and preoperative echocardiograms of consecutive patients who underwent aortic or mitral valve surgery from 1986 to 1994 at the Cleveland Clinic Foundation were reviewed. The study group consisted of 1,985 patients, 1,476 with aortic stenosis and 509 with mitral stenosis. Preoperative cardiac catheterization was performed in 1,456 patients with aortic stenosis (99%) and 488 with mitral stenosis (96%). Measurement of invasive hemodynamics (including transvalvular gradients and estimated valve areas) during catheterization decreased over time both in patients with aortic (from 64% in 1986 to 30% in 1994, test for trend p <0.0001) and mitral stenosis (from 63% in 1986 to 18% in 1994, test for trend p <0.0001). After adjusting for age, gender, and other characteristics, the only predictors of performance of invasive hemodynamics in patients with aortic stenosis were more recent surgery (inverse relation, p = 0.0001) and New York Heart Association class (p = 0.01); in patients with mitral stenosis the only predictor was also more recent surgery (inverse relation, p = 0.0001). Thus, use of preoperative invasive hemodynamics in patients with valvular stenosis has markedly decreased over the last decade. This is an example of how a noninvasive modality can supercede an invasive one, even when surrounding a procedure as fundamentally invasive as valvular heart surgery.
OBJECTIVES: This study sought to determine whether left atrial appendage stunning occurs in patients with atrial flutter and to compare left atrial appendage function in the pericardioversion period with that in patients with atrial fibrillation. BACKGROUND: Left atrial appendage stunning has recently been proposed as a key mechanistic phenomenon in the etiology of postcardioversion thromboembolic events in atrial fibrillation. Atrial flutter is thought to be associated with a negligible risk of thromboembolic events; therefore, anticoagulation is commonly withheld before and after cardioversion in these patients. METHODS: Sixty-three patients with atrial flutter (n = 19) or atrial fibrillation (n = 44) underwent transesophageal echocardiography immediately before and after electrical cardioversion. In addition to assessing the presence of thrombus and spontaneous echo contrast, we measured left atrial appendage emptying velocity and calculated shear rates by pulsed wave Doppler and two-dimensional echocardiography. RESULTS: Patients with atrial flutter exhibited greater left atrial appendage flow velocities before cardioversion than those with atrial fibrillation (42 +/- 19 vs. 28 +/- 15 cm/s [mean +/- SD], p < 0.001). Left atrial appendage shear rates were also higher in patients with atrial flutter (103 +/- 82 vs. 59 +/- 37 s-1, p < 0.001). After cardioversion, left atrial appendage flow velocities decreased compared with precardioversion values in patients with atrial fibrillation (28 +/- 15 before to 15 +/- 14 cm/s after cardioversion, p < 0.001) and atrial flutter (42 +/- 19 to 27 +/- 18 cm/s, respectively, p < 0.001). Shear rates decreased from 59 +/- 37 before cardioversion to 30 +/- 31 s-1 after cardioversion in atrial fibrillation (p < 0.001), and from 103 +/- 82 s to 65 +/- 52 s-1, respectively (p < 0.001), in atrial flutter. This decrease in flow velocity from before to after cardioversion occurred in 36 (82%) of 44 patients with atrial fibrillation and 14 (74%) of 19 with atrial flutter. The impaired left atrial appendage function after cardioversion was less pronounced in the group with atrial flutter (27 +/- 18 cm/s for atrial flutter vs. 15 +/- 14 cm/s for atrial fibrillation, p < 0.001). New or increased spontaneous echo contrast occurred in 22 (50%) of 44 patients with atrial fibrillation versus 4 (21%) of 19 with atrial flutter (p < 0.05). CONCLUSIONS: Left atrial appendage stunning also occurs in patients with atrial flutter, although to a lesser degree than in those with atrial fibrillation. These data suggest that patients with atrial flutter are at risk for thromboembolic events after cardioversion, although this risk is most likely lower than that in patients with atrial fibrillation because of better preserved left atrial appendage function.
BACKGROUND: A new Doppler echocardiographic technique has been developed for automated cardiac output measurement (ACOM) that assumes neither a flat flow profile nor collinearity with the scan line, but clinical validation of this method is lacking. METHODS AND RESULTS: In 165 subjects (50 intensive care patients, 10 dobutamine echocardiography patients, and 105 normal volunteers; age, 49.4 +/- 19.3 years; 92 men), ACOM was performed in the left ventricular outflow tract (LVOT), with the color baseline shifted to avoid aliasing. ACOM was also tested in a pulsatile in vitro model. Stroke volume was calculated by double integration of Doppler signals in space (across the LVOT) and in time (through the systolic period), assuming hemiaxial symmetry: integral of integral of pi r v(r,t) dr dt, where v(r,t) is the velocity at a distance r from the center of the LVOT at time t during systole. Stroke volume from ACOM was compared with thermodilution (TD), aortic valve pulsed-wave Doppler (PWAO), and left ventricular echocardiographic (two-dimensional [2D]) methods. There was good correlation between ACOM and PWAO (r = .93). TD (r = .86), and 2D (r = .74), with close agreement seen. ACOM had higher correlation and agreement with TD than did either PWAO (P < .02) or 2D (P < .01). ACOM was also able to track accurately the changes in cardiac output with dobutamine infusion in comparison with PWAO (r = .94). In vitro assessment demonstrated excellent correlation (r = .98, y = 1.0x + 1.94) with little impact of pulse repetition frequency or misalignment up to 30 degrees. Gain dependency was noted but could be optimized by visual inspection of the color image. CONCLUSIONS: Automatic integration of numerical data within color Doppler flow fields is a feasible new method for quantifying flow. It is simpler and faster, requires fewer assumptions, and uses only one apical view. ACOM is a promising new approach to echocardiographic quantification that deserves further study and refinement.
BACKGROUND: Intraoperative echocardiography is a valuable monitoring and diagnostic technology used in cardiac surgery. This reports our clinical study of the usefulness of intraoperative echocardiography to both surgeons and anesthesiologists for high-risk coronary artery bypass grafting. METHODS: From March to November 1995, 82 consecutive high-risk patients undergoing coronary artery bypass grafting were studied in a four-stage protocol to determine the efficacy of intraoperative echocardiography in management planning. Alterations in surgical and anesthetic/hemodynamic management initiated by intraoperative echocardiography findings were documented in addition to perioperative morbidity and mortality. RESULTS: Intraoperative echocardiography initiated at least one major surgical management alteration in 27 patients (33%) and at least one major anesthetic/hemodynamic change in 42 (51%). Mortality and the rate of myocardial infarction in this consecutive high-risk study population using intraoperative echocardiography and in a similar group of patients without the use of intraoperative echocardiography was 1.2% versus 3.8% (not significant) and 1.2% versus 3.5% (not significant), respectively. CONCLUSIONS: We conclude that when all of the isolated diagnostic and monitoring applications of perioperative echocardiography are routinely and systematically performed together, it is a safe and viable tool that significantly affects the decision-making process in the intraoperative care of high-risk patients undergoing primary isolated coronary artery bypass grafting and may contribute to the optimal care of these patients.
OBJECTIVES: We evaluated the value of preoperative assessment of left ventricular contractile reserve in predicting ventricular function after valve repair for minimally symptomatic mitral regurgitation. BACKGROUND: The optimal timing for operation in minimally symptomatic patients with significant mitral regurgitation is controversial. Accurate preoperative assessment of left ventricular function is difficult, and the ability to predict postoperative function is limited. Previous studies in patients undergoing mitral valve replacement may not be applicable in the present era of valve repair. METHODS: We performed exercise echocardiography in 139 patients with isolated mitral regurgitation and no coronary disease, 74 of whom subsequently underwent uncomplicated valve repair. We measured rest left ventricular end-systolic dimension, end-systolic wall stress and positive first derivative of left ventricular pressure (dP/dt). End-diastolic and end-systolic volumes and ejection fraction were measured preoperatively at rest, immediately after exercise and postoperatively. RESULTS: Ejection fraction decreased postoperatively to 55 +/- 10% from a rest preoperative value of 64 +/- 9% (p < 0.001). Compared with patients with a postoperative ejection fraction > or = (n = 56), patients with postoperative ejection fraction < 50% (n = 18) had a significantly lower preoperative exercise ejection fraction (57 +/- 11% vs. 73 +/- 9%, p < 0.0005), a larger exercise end-systolic volume index (32 +/- 8 vs. 18 +/- 7 cm3/m2, p < 0.0005) and a lower change in ejection fraction with exercise (-4 +/- 8% vs. 9 +/- 10%, p < 0.005). Preoperative rest indexes, including dP/dt, end-systolic wall stress and end-systolic volume index were less predictive, whereas exercise capacity, rest ejection fraction and end-systolic dimension were not predictive of post-repair ejection fraction. An exercise end-systolic volume index > 25 cm3/m2 was the best predictor of postoperative dysfunction, with a sensitivity and specificity of 83%. CONCLUSIONS: In minimally symptomatic patients with mitral regurgitation, latent ventricular dysfunction may be indicated by a limited contractile reserve, manifest at exercise as an inadequate increase in ejection fraction and a larger end-systolic volume. These variables may also be used to predict left ventricular function after repair.
OBJECTIVES: The aims of this study were to define the hydrodynamic mechanisms involved in the occurrence of hemolysis in prosthetic mitral valve regurgitation and to reproduce them in a numeric simulation model in order to estimate peak shear stress. BACKGROUND: Although in vitro studies have demonstrated that shear stresses > 3,000 dynes/cm2 are associated with significant erythrocyte destruction, it is not known whether these values can occur in vivo in conditions of abnormal prosthetic regurgitant flow. METHODS: We studied 27 patients undergoing reoperation for significant mitral prosthetic regurgitation, 16 with and 11 without hemolysis. We classified the origin and geometry of the regurgitant jets by using transesophageal echocardiography. By using the physical and morphologic characteristics defined, several hydrodynamic patterns were simulated numerically to determine shear rates. RESULTS: Eight (50%) of the 16 patients with hemolysis had paravalvular leaks and the other 8 had a jet with central origin, in contrast to 2 (18%) and 9 (82%), respectively, of the 11 patients without hemolysis (p = 0.12, power 0.38). Patients with hemolysis had patterns of flow fragmentation (n = 2), collision (n = 11) or rapid acceleration (n = 3), whereas those without hemolysis had either free jets (n = 7) or slow deceleration (n = 4) (p < 0.001, power 0.99). Numeric simulation demonstrated peak shear rates of 6,000, 4,500, 4,500, 925 and 950 dynes/cm2 in these five models, respectively. CONCLUSIONS: The distinct patterns of regurgitant flow seen in these patients with mitral prosthetic hemolysis were associated with rapid acceleration and deceleration or high peak shear rates, or both. The nature of the flow disturbance produced by the prosthetic regurgitant lesion and the resultant increase in shear stress are more important than the site of origin of the flow disturbance in producing clinical hemolysis.
To validate the use of pulsed Doppler tissue imaging that measures myocardial wall velocities and to define the characteristics of these velocities in normal subjects, we obtained and compared the anteroseptal and posterior wall velocities in 24 volunteers with pulsed Doppler tissue imaging and digitized M-mode echocardiography. We also studied the relation between velocity components and hemodynamic events timed by standard Doppler flows. There was an excellent correlation between Doppler and M-mode-derived velocities (r = 0.95, p < 0.001), with higher reproducibility for Doppler (r = 0.99) than for M-mode (r = 0.95, p < 0.001). Biphasic velocities that were uniformly present during isovolumic contraction and relaxation were attributed to geometric changes due to asynchronous contraction and ventricular interdependence. We conclude that wall velocities obtained by pulsed Doppler tissue imaging are accurate and reproducible. This method may prove useful for studying the contractile and elastic properties of the myocardium.
To ascertain the factors related to the transesophageal echocardiographic diagnosis of patent foramen ovale, the primary echocardiographic data were reviewed from 74 patients with that diagnosis. Similar detection rates were observed with contrast echocardiography (82%) and color flow mapping (78%), but contrast was more sensitive in patients with normal-sized atria (94%) and isolated right atrial enlargement (100%) than in those with enlarged left atria, all but three of whom had biatrial enlargement (52%) (p < 0.001 by chi-square test). Doppler flow mapping was less sensitive for normal-sized atria (66%) than for isolated enlarged right atria (81%) and left (bi-) atrial enlargement (96%) (p = 0.024). Patent foramen ovale was directly visualized by two-dimensional echocardiography in 80%, with close agreement to the size of the color flow jet (r = 0.90, delta = 0.1 +/- 0.4 mm); foramen ovale size was related to right atrial area (r = 0.31, p = 0.015). Thus both saline contrast and color flow imaging are necessary to exclude the diagnosis of patent foramen ovale.
The aim of this study was to characterize left atrial appendage mechanical function in atrial fibrillation and flutter by Fourier analysis to analyze frequency and regularity of flow. Left atrial appendage function is central to a patient's risk for thromboembolism. Although the function of the appendage can be analyzed by Doppler echocardiography in sinus rhythm, its mechanical function in atrial fibrillation and flutter has not been well characterized. This lack of adequate definition is caused by the complexity and temporal variability of the Doppler flow profiles. We assessed left atrial appendage function in 21 cases of atrial fibrillation (n - 11) and flutter (n = 10) and five in sinus rhythm with transesophageal Doppler echocardiography. Doppler profiles were examined by Fourier analysis, and the power spectra compared and analyzed between patients with atrial fibrillation and flutter. Left atrial appendage Doppler flow in atrial fibrillation produced Fourier spectra over a narrow band of frequencies with a peak frequency of 6.2 +/- 1.0 Hz, significantly higher than in atrial flutter (3.9 +/- 0.6 Hz, p < 0.00001). Additionally, a significant difference in subharmonic modulation (spectral power below the peak frequency) was observed between atrial appendage flow in atrial fibrillation and flutter, because 37% +/- 16% of the total spectral power was achieved before the dominant frequency in atrial fibrillation compared with 20% +/- 14% in atrial flutter (p = 0.02). Conversely, patients in sinus rhythm exhibited broad-banded Fourier spectra with most of the power in discrete frequency spikes at harmonics above the fundamental frequency with very little subharmonic modulation (1% +/- 0.05%). Left atrial appendage function in atrial fibrillation and flutter can be well characterized by Fourier analysis of Doppler flow. Atrial fibrillation has higher dominant frequencies and greater subharmonic modulation compared with flutter. Moreover, atrial fibrillation demonstrated quasiperiodic contraction patterns typically found in chaotic systems. Fourier analysis of left atrial appendage contraction patterns may therefore have significant promise in providing insights into mechanisms of atrial fibrillation and thromboembolism.