YEAR IN CARDIOLOGY SERIES
The Year in Echocardiography
Arthur E. Weyman, MD*
Massachusetts General Hospital, Boston, Massachusetts.
Manuscript received November 9, 2005;
accepted December 1, 2005.
* Reprint requests and correspondence: Dr. Arthur E. Weyman, Cardiac Ultrasound Laboratory, 55 Fruit Street, YAW 5, Boston, Massachusetts 02114. (Email: aweyman{at}partners.org).
Each year echocardiography progresses along similar paths. These paths consist of: 1) the introduction and validation of new technology; 2) the refinement and extension of existing diagnostic methods; and 3) the application of new and established techniques to areas of clinical focus. Over time, however, the introduction of new techniques in any area gradually decreases, and in areas where analysis is complex and time-consuming, the focus shifts to more rapid and automated processing and quantitation. So it has been in the past year in echocardiography, where the major advances have been in the areas of image and Doppler processing and automated analysis. In this review, an attempt is made to highlight and clarify some of these new methods and to indicate areas of potential clinical application. However, in analyzing these data and methods, it is important to remember that the more automated a process, the further removed the operator becomes from the raw data and hence the underlying sources of error. Understanding these potential sources of error therefore becomes increasingly important. It is also important to distinguish between techniques that inherently answer important questions and those searching for a distinguishing question to answer.
 |
Introduction and validation of new technology
|
|---|
Real-time 3-D echocardiography (RT3-DE).
RT3-DE imaging and Doppler devices have been available for several years. Unfortunately, limited image resolution and lengthy analysis time have limited widespread application of RT3-DE, and the incremental value of this approach in clinical practice is still being defined. Most research to date has focused on the validation of RT3-DE for the assessment of global and local ventricular function, with feasibility documented and correlations with objective standards generally superior to those obtained using conventional two-dimensional methods when volumes were obtained at end diastole and end systole. More recently, the ability of a semi-automated RT3-DE approach to provide continuous measures of global and regional left ventricular (LV) volume, volume change, and regional time-wall motion curves was assessed in comparison with cine magnetic resonance imaging (MRI). The authors used a level set approach to identify the endocardial surface after first guiding the algorithm by manually defining multiple points along the endocardial surface from equiangular, end-diastolic apical planes separated by 45°. From this starting point, the algorithm identified (using the preceding frame as an initial condition) the endocardial surfaces of sequential frames for the entire cardiac cycle. The LV long axis (a line drawn from the midpoint of the mitral annulus to the apical tip) was then trisected; six wedge-shaped segments for each level (base, mid-ventricle, and apex) were defined; and the total volume, volume at each level, and volume for each segment were computed for each point in the cardiac cycle. Total analysis time was <15 min. In a group of 16 patients selected for adequate two-dimensional images, linear regression analysis for RT3-DE and MRI end-diastolic volume (EDV), end-systolic volume (ESV), and ejection fraction (EF) yielded high correlation coefficients (r = 0.99, 0.99, and 0.98, respectively) with regression slopes near unity. When point-by-point volumes were compared, the overall correlation was again excellent (r = 0.99). However, the 95% confidence intervals (CIs) increased to ±20 ml (18.7% of the mean MRI value). When partial volumes were analyzed, the agreement at the base and mid-ventricle was maintained, but decreased to r = 0.80 for the apical segments. The correlation between MRI- and RT3-DE–derived peak wall motion was modest (r = 0.69) when all segments were combined with 95% CIs of ±4.6 mm. When apical segments were removed, the correlations improved (r = 0.89) with the 95% CI of ±2.6 mm. When patients with global LV dysfunction (dilated cardiomyopathy n = 9) were compared with a similar group of selected normal subjects, RT3-DE showed significant differences between group difference in EDV, ESV, EF, peak ejection rate, peak rapid filling rate, peak atrial filling rate, index of systolic asynchrony (ISA) (index of LV systolic asynchrony = the standard deviation of the time intervals from the R-wave of the electrocardiogram [ECG] to the minimal regional volume), and the regional shortening fraction (all p < 0.05). Finally, using the regional fractional shortening data from normal subjects, it was possible to automatically classify 170 of 198 segments (86%) as normal or abnormal in 11 patients with coronary artery disease when compared to expert analysis of two-dimensional images. Thus, while this approach offers the advantage of being largely automated, the analysis time remains long for clinical application, and, as the authors note, using a floating center diminishes the effects of translation, albeit at the expense of artificially augmenting reported contraction in dysfunctional segments opposite normally functioning areas. Although this technique is promising, much larger studies in non-selected groups will be required to establish its ultimate clinical value (1).
RT3-DE was also used to estimate regional dyssynchrony in normal subjects and in patients with varying degrees of LV dysfunction. Regional LV function was assessed from time-volume curves of 16 pyramidal volumes originating from a computed center of volume, with bases corresponding to segments in a 16-segment LV model. A segmental dyssynchrony index (SDI) was defined from the standard deviations of the times taken for each segment to reach minimal volume, expressed as a percentage of the cardiac cycle, with higher SDIs indicating a greater degree of dyssynchrony. An arbitrary cutoff of three standard deviations was taken to indicate mechanical dyssynchrony. Normal patients had a mean SDI of 3.5 ± 1.8%, compared with 4.5 ± 2.4% for patients with cardiovascular disease and a normal EF, and 5.4 ± 0.8%, 10.8 ± 2%, and 15.6% ± 1% for patients with mild and moderate/severe LV dysfunction, respectively (p for trend < 0.001). The correlation between LV systolic function and SDI persisted irrespective of QRS duration (r = 0.79 and r = 0.77 for LVEF in patients with broad [>120 ms] and narrow [<120 ms] QRS duration, respectively). In contrast, there was only a weak correlation between QRS duration and SDI (r = 0.284). Surprisingly, dyssynchrony was more common in patients with moderate and severe LV dysfunction and QRS intervals <120 ms than in those with similar dysfunction and QRS intervals >120 ms (78.8% vs. 62.0%), raising questions about the specificity of the method. Cardiac resynchronization therapy (CRT) was performed in a subset of 26 patients (QRS  130 ms), with 23 available for follow-up at 10 ± 1 months. Responders (n = 17) had higher SDIs before the procedure than non-responders with similar EFs, and demonstrated a significant decrease in dyssynchrony and an increase in LVEF. Despite these promising results, there are a number of methodologic issues that must be considered, and, as the authors point out, the accuracy of the method has never been documented relative to a gold standard. The potential sources of error include: 1) the calculation of the centroid (method not described) and delineation of the segment boundaries, which must be recomputed for each frame and which can independently influence the volume and change in volume of any segment, since the instantaneous volume of one segment is influenced by all the others; 2) errors in border definition by the automated border detection algorithm, which are always present; 3) the limited sampling rate and smoothing of the volume time curves, which can affect the perceived time of minimal volume, and 4) registration errors due to reconstruction of instantaneous volumes from four cardiac cycles. Errors in volume calculation may not be a problem in normal ventricles, where the interframe differences are great, but as LV function deteriorates and absolute volume changes become very small, errors introduced in the analysis process will be magnified. The use of the dyssynchrony index masks these errors by separating the result from the underlying data. Finally, counting indexes (total number of regions with post-systolic dysfunction) and variance parameters (statistical variation of time to maximal motion) fail to incorporate the spatial placement of these regions. This appears important since the net mechanical deficiency from discoordinate contraction is greater when segments that are out of phase are geographically clustered rather than being dispersed throughout the wall (2). Alternatively, the three-dimensional approach may simply be more sensitive because it measures radial motion rather than longitudinal velocity and circumferential strain (closely linked to radial strain/contraction), which has recently been shown by MRI tagging studies in a canine heart failure model to be more sensitive to mechanical dyssynchrony, to follow a different time course, and to provide different information about response to CRT (3).
The ability of RT3-DE with color Doppler to measure cardiac output (CO) was also assessed in an experimental study by comparison with outputs measured by an ultrasonic flow probe. A color Doppler plane at the level of the aortic valve was first defined, and then the outflow area and integrated mean velocity within that area were determined. There was good correlation between CO derived from the three-dimensional Doppler and flow probe data (r2 = 0.93) that improved when the Doppler and flow probe data were averaged. The RT3-DE overestimated flow probe CO by 7.3%. The acquisition was facilitated by the use of small animals (2.7 to 11.3 kg), which permitted a relatively high frame rate and Nyquist limit and direct epicardial probe placement, so determination of clinical utility again awaits further study (4).
The RT3-DE has also been reported to be of value for measuring stroke volume (5), LV mass (6), atrial defect size (7), LV ESV, EDV, mass, stroke volume, and EF in children (8), as well as left atrial volume (9), myocardial perfusion (10), quantitation of mitral valve tenting in ischemic MR (11), and tricuspid stenosis (12). In addition, a number of reports have appeared describing the utility of RT3-DE in individual patients. Despite this steady progress, however, routine clinical application remains a goal rather than a reality. While RT3-DE studies, recapitulating the results from earlier reports based on the reconstruction of spatially located planes, have shown that three-dimensional methods are more accurate and reproducible than two-dimensional echocardiography for the measurement of cardiac chamber size and morphology, analysis time continues to limit clinical application.
Tissue/speckle tracking.
During the past several years, Doppler tissue imaging (DTI) has proven to be a valuable technique for measuring global and local ventricular function. The technique is limited, however, in that it can only reliably report function along the axis of the ultrasound beam, rather than describing the more complex two- and three-dimensional patterns of cardiac motion and deformation that more completely describe cardiac function. To overcome these problems, several tissue-tracking algorithms have been developed that attempt to map the motion of the myocardium directly from the radio frequency data that form the basis for two-dimensional B-mode ultrasound scans. Although they differ slightly, these approaches basically begin by defining a pixel within a search range placed within the myocardium and then mapping the pattern of intensity amplitudes (mask) in a given region around that pixel. Moving to the next frame, the algorithm begins at the same initial pixel location and then sweeps the pattern or mask from the preceding frame in a pre-defined search pattern around that pixel until the optimal correlation between the mask and underlying data is obtained. Using this approach, the displacement of the mask from its starting point, in theory, represents the interframe movement of the underlying tissue. Repeating the processes from frame to frame throughout the cardiac cycle at multiple points around the ventricle gives the amplitude, rate, and direction of motion at all of the selected starting points and hence permits the calculation of direction, and velocity of motion and two-dimensional strain. This approach has the additional advantage of being independent of in-plane translation and recording angle. However, there are several potential limitations. First, the intra-myocardial echocardiograms (speckle) being tracked are generally considered to be a random phenomenon due to the constructive and destructive interference of the acoustic waves reflected by scatterers in the myocardium (structures that are smaller than wavelength), rather than arising from specific structures; therefore, their consistency between frames is not assured and, when present, correlation is generally possible for only short time intervals. Second, when studying fixed short-axis planes, apex-to-base shortening during systole will cause the portion of the myocardium imaged during each frame to be slightly different, so that the speckle pattern should also change independently and the accuracy of the method may be affected by the contraction of the ventricle. Finally, echocardiographic images are inherently noisy, and noise that is random can interfere with the interframe correlations. As a result, required correlation thresholds are introduced, below which data are considered unreliable and hence excluded. The technique would seem to be most reliable in conditions where the myocardium is thick and where image quality is good.
Early validation studies have been based on small numbers of patients with primarily aortic valve disease and, presumably, LV hypertrophy, in which speckle is more obvious and may arise in part from fixed internal structures and hence be less variable. In one such study of 15 patients, a speckle-tracking algorithm was compared to DTI velocity and tagged MRI studies for the measurement of rotational motion and velocity. Tissue tracking was feasible in 13 of 15 patients, with very good correlation for both LV rotation and rotational velocity between the speckle-tracking algorithm and both of the reference techniques (13).
Using a similar approach in an open-chest ovine model, longitudinal and radial strain calculated using a tissue-tracking algorithm were compared with sonomicrometer-measured values, with good correlations obtained (r = 0.72 and 0.80) for radial and longitudinal strain, respectively (14).
Multiscale motion mapping (MMM) is another tissue-tracking approach that similarly uses the grayscale levels of individual pixels input into the affine motion equation that fits best to individual regions to predict local translation, rotation, and deformation. The approach is attractive because it offers various display formats depicting local velocity vectors and both the radial and circumferential components of strain simultaneously (only short-axis views were demonstrated). The approach has performed well in tissue phantoms, and when compared with tissue Doppler from 11 clinical echocardiograms, there was a good correlation between MMM and DTI (r = 0.94) with a standard deviation of 7.2% of the measured range. In 114 clinical cases, wall motion classified as normal or abnormal by MMM corresponded to a "majority vote" of clinical experts in 84% of cases. Further clinical validation is in progress (15).
|
Refinement and extension of existing diagnostic method
|
|---|
Contrast echocardiography.
Detection of coronary stenosis at rest
Current myocardial contrast echocardiographic (MCE) methods for quantitating myocardial blood flow (MBF) rely on the destruction and measurement of the rate of subsequent replenishment of contrast within the region of the myocardium transected by the ultrasound beam. Because 90% of the myocardial blood volume (MBV) resides in the capillaries and capillary flow rates approximate 1 mm/s, it takes  4 to 5 s to replenish the myocardium insonated by a beam with an elevation dimension of 4 to 5 mm. However, part of this replenishment occurs from microbubbles moving at faster velocities in larger intramyocardial vessels, such as arterioles. The pulsing interval versus microbubble backscatter amplitude during replenishment, therefore, is curvilinear, with faster velocities depicting more rapid arteriolar filling. As blood in myocardial arterioles represents only a small portion of the MBV, backscatter from these vessels is usually negligible when the beam is fully replenished. However, if the interval between destructive pulses is short, the signal obtained arises only from the arterioles, since neither the capillaries nor the venules have time to fill. Thus, the signal representing arteriolar blood volume (aBV) can, in theory, be obtained by this approach. As forward flow in large intramyocardial vessels occurs predominantly in diastole, aBV signals on MCE are normally seen predominantly during diastole.
However, in the presence of coronary stenosis, autoregulation causes dilation of 150- to 300-µm arterioles, resulting in an increase in arteriolar and total coronary blood volume that increases as the degree of coronary stenosis increases. During systole the change in intracavitary pressure is transmitted to the subendocardium, causing retrograde displacement of the aBV into the larger vessels, resulting in a systolic signal on MCE. Because aBV is larger when a stenosis is present, there is likely to be a greater inflow of microbubbles from adjacent smaller arterioles and an increase in the systolic signal. A prior animal study demonstrated that that the systolic-to-diastolic (S/D) aBV signal ratio measured at rest increases in proportion to the degree of coronary stenosis (16). In an attempt to see if these findings could be reproduced in the clinical environment, patients with varying degrees of coronary stenosis on quantitative angiography underwent high-mechanical-index MCE at 15 Hz to allow measurement of phasic changes in aBV in large intramyocardial vessels. Patients were studied using either Definity (n = 22, group 1) or Imagent (n = 22, group 2). For each level of stenosis severity (none, mild [<50%], moderate [50% to 75%], and severe [>75%]), progressive increases in the background subtracted systolic/diastolic aBV signal ratio were noted: group 1 (0.09 ± 0.13, 0.13 ± 0.08, 0.58 ± 0.22, and 0.77 ± 0.40, p < 0.001) and group 2 (0.10 ± 0.05, 0.27 ± 0.18, 0.39 ± 0.28, and 0.74 ± 0.37, p < 0.0001). In group 1 patients, an S/D aBV ratio of >0.34 provided sensitivity and specificity of 80% and 71%, respectively, for the detection of >75% stenosis, whereas as ratio of >0.43 provided a sensitivity and specificity of 89% and 74% in group 2 patients. This study demonstrates for the first time that coronary stenoses can be detected by MCE at rest in patients with suspected coronary disease. However, the authors suggest that the ratio may be independently affected by heart rate, cardiac contractility, and collateral flow. In addition, systolic arteriolar flow reversal is not uniform throughout the myocardium, which may further complicate the use of this technique (17,18).
In other studies, it was reported that: 1) MCE permitted differentiation of ischemic from non-ischemic acute heart failure on the basis of the detection of flow-limiting stenoses (19); 2) when combined with regional function assessment, MCE improved early risk assessment in a large group of patients presenting to the emergency department with chest pain and non-diagnostic ECGs—potentially improving triage—and when combined with clinical variables, was superior to the Thrombolysis In Myocardial Infarction (TIMI) score in predicting subsequent long-term events (20); and 3) when combined with dobutamine-atropine stress echocardiography (DSE), MCE and wall-motion abnormality were equally sensitive for detecting left anterior descending coronary artery stenosis (>70%) (however, MCE increased the sensitivity of detection of right coronary artery and circumflex coronary stenoses. Although in all cases wall-motion abnormalities were more specific, MCE was also more sensitive for detecting multivessel disease with only a modest loss of specificity [21]); and 4) when added to DSE, MCE also had a significant incremental prognostic value over clinical factors, EF, and wall-motion response to exercise in predicting subsequent death or non-fatal myocardial infarction. The three-year event-free survival rates were 95% for patients with normal wall motion and myocardial perfusion, 82% for normal wall motion and abnormal myocardial perfusion, and 68% for abnormal wall motion and myocardial perfusion (22). In each study, the incremental value of MCE was attributed to its ability to detect perfusion abnormalities and predict underlying coronary stenoses before the development of segmental dysfunction.
Doppler tissue imaging.
Prior experimental studies have suggested that isovolumic myocardial acceleration (IVA) by DTI represents a relatively load-insensitive index of LV contractility. In a study designed to examine the role of IVA in assessing regional function, IVA was assessed by TDI, which measures net motion in a region of interest and by sonomicrometry, which measures the local segmental contribution to motion during changes in loading conditions, contractility, and ischemia. Dobutamine increased contractility, as assessed by dP/dtmax, with parallel increases in isovolumic contraction (IVC) velocity and IVA by sonomicrometers and DTI. In contrast, volume loading caused a small increase in contractility, but IVA decreased markedly. The response of IVA to changes in pre-load was biphasic, with IVA decreasing following both a decrease in pre-load by IVC occlusion and an increase due to volume loading. Likewise, during various degrees of ischemia, there was a progressive decrease in systolic shortening, but IVA was preserved. Thus, it appears that while IVA parallels the global response to increasing contractility by dobutamine infusion, it does not reflect regional contractility and is pre-load dependent (23).
To test the clinical utility of DTI for differentiating transmural from non-transmural infarction, 47 consecutive patients with acute myocardial infarction were studied 2 to 6 days after admission by DTI/strain-rate imaging (SRI) and contrast-enhanced magnetic resonance imaging (Ce-MRI). Using receiver-operating characteristic (ROC) analysis, a peak systolic strain rate of >–0.59 s–1 separated transmural from non-transmural and subendocardial myocardial infarction, with a sensitivity of 90.9% and 90.9% and specificity of 96.4% and 100%, respectively. A cutoff value of –0.98 s–1 systolic strain rate >–1.26 s–1 was able to distinguish subendocardial infarction from normal with a sensitivity of 81.3% and a specificity of 83.3%. Thus, although Ce-MRI remains the gold standard, the wider availability of SRI could make it a useful alternative (24).
Dobutamine stress echocardiography.
A restrictive filling pattern (RFP) as determined by Doppler echocardiography in patients with dilated cardiomyopathy is associated with severe impairment of LV hemodynamics and is a powerful predictor of increased mortality. In a study examining whether persistence of an RFP at peak dobutamine stress had additional predictive value, 69 patients with ischemic cardiomyopathy were studied at rest and at peak stress. Forty-two patients had an RFP at rest. At peak stress, the RFP normalized in 24 but persisted in 18 patients. Three-year survival rates for patients with no RFP, a resting RFP that normalized at peak exercise, and a persistent RFP were 89%, 79%, and 49%, respectively (p < 0.001). Compared with no RFP, the hazard ratios (HRs) for reversible and irreversible RFP were 1.9 and 9.5, respectively. The presence of a persistent RFP was felt to imply a rise in left atrial (LA) pressure and blunted inotropic response that identified high-risk patients.
Heart failure.
Prognosis
Although a wide variety of echocardiographic parameters can be measured or calculated in patients with heart failure, many are physiologically or mathematically related, and the relative prognostic value of these variables has been unclear. In early heart failure trials, LV size and EF were shown to predict outcome; however, these failed to include Doppler variables or the degree of mitral regurgitation. Analysis of imaging and Doppler data from 336 heart failure patients (mean EF 24.9%) in the echocardiography substudy of the Beta-blocker Evaluation of Survival Trial (BEST) showed that although a number of imaging and Doppler parameters predicted death or a combined end point of death, hospitalization for heart failure, or transplantation by univariate analysis, only LV EDV index (>120 ml/m2) was predictive of death, and only an LV EDV index of >120 ml/m2), a mitral deceleration time (DT)  150 ms, and a vena contracta width >0.4 cm were predictive of the combined end point after adjustment for clinical covariates. Although LV EF was predictive of survival on univariate analysis, it was not predictive on multivariate analysis, likely reflecting the limited range of EFs in this study (18). Because of its date of initiation, the study failed to include tissue Doppler or SRI.
The incremental role of tissue Doppler in predicting outcome was examined in a study of 193 patients with EFs <45%. An RFP (defined as an E/A ratio >2, E' <8 cm/s, and a DT <150 ms) was the strongest multivariate predictor of a subsequent cardiac event (HR = 6.62) during a mean follow-up of 1 year. Other multivariate predictors included QRS duration >144 ms (HR = 4.26), LV systolic dimension index >2.75 cm/m2 (HR = 3.34), and E' <5.5 cm/s (HR = 2.48). The LV EF was a univariate but not a multivariate predictor (25).
In a longer follow-up study of 182 patients with impaired systolic function (EF <50%; mean 36 ± 9%), the predictive value of mitral annulus motion (average of four basal segments) by DTI for cardiac death during a median follow-up of 48 months was examined. An early diastolic mitral annular velocity (Em <3) remained the strongest predictor for cardiac death in a multivariate Cox regression backward analysis adjusted by age, Sm, Em, Am, E/Em, and DT. In a model containing clinical risk factors, DT 140 and E/Em, addition of Em <3 significantly improved the predictive value of the model. In this less severely affected population, neither moderate/severe mitral regurgitation nor LV systolic dimension was predictive of outcome. Thus parameters of early diastolic filling (DT and E') appear to be more predictive of outcome than EF in patients with primarily systolic dysfunction, although the appropriate cutoff values and application in different degrees of failure still need to be defined (26).
Treatment—CRT
The role of echocardiography methods in identifying patients with advanced heart failure most likely to benefit from biventricular pacing (BVP), determining the optimal pacing sequences for individual patients, and measuring the time course and degree of function/morphologic response continues to be an area of significant interest. In a detailed study of 41 patients with heart failure undergoing BVP, cardiac output, severity of mitral regurgitation, and LV filling time and parameters of inter- and intraventricular synchrony were evaluated by echocardiography during a variety of pre-determined pacing modes, including: 1) RV pacing, 2) LV pacing, 3) simultaneous BVP pacing, 4) sequential BVP with RV pre-activation at interventricular intervals of 12, 20, 40, and 80 ms, and 5) sequential BVP with LV pre-excitation at similar interventricular intervals. Tissue Doppler was used to define the time to the onset and peak of ventricular contraction for six basal and six mid-ventricular segments, with the intra-LV peak difference equal to the difference between the shortest and longest of the 12 electromechanical delay values. An index of systolic dyssychrony was also calculated as the standard deviation of the time to peak segmental contraction within each segment. Delayed contraction was also defined by strain rate analysis when the peak segmental contraction occurred after aortic valve closure, and summed for affected segments. Compared to baseline, BVP increased cardiac output and LV filling time and decreased effective mitral regurgitant orifice area (EROA). Optimal BVP, defined as the pacing sequence that produced the maximum increase in cardiac output, resulted in a further increase in mean filling period and decrease in EROA. Changes in the degree of asynchrony observed during the various pacing modes correlated with changes in cardiac output. These correlations were highly significant for intra-LV delay onset (r = –0.64, p < 0.001), intra-LV contraction peak (r = –0.67, p < 0.001), and index of LV dyssynchrony (r = –0.67, p < 0.001), whereas septal-posterior wall-motion delay (r = –0.41, p < 0.05) and extent of myocardial DLC changes (r = 1 0.41, p < 0.05) revealed significant but lower correlations. This study helps define the mechanism of improvement following BVP by demonstrating a strong correlation between improvement in cardiac output and parameters of ventricular dyssynchrony. The study also suggests, but fails to prove, that optimized BVP as compared to BVP may reduce the number of non-responders to CRT (27).
Because of the time required to analyze tissue Doppler data, new computer-based analysis programs are continually being developed in the hope of streamlining this process. One such approach is tissue synchronization imaging (TSI), which portrays regional asynchrony on two-dimensional echocardiographic images by transforming the timing of regional peak velocity into color codes, which in theory allow immediate visual identification of regional delay in systole by comparing color mapping of orthogonal walls qualitatively or quantitatively. In a study of 56 heart failure patients who received CRT, the utility of TSI to predict reverse remodeling was examined. The most common site of the most severe delay was the inferior wall (25 of 56 patients, 45%), followed by the lateral wall (30%), posterior wall (25%), septal wall (16%), anterior wall (11%), and anteroseptal wall (5%). Only the presence of lateral wall delay at baseline was predictive of a positive reverse remodeling effect after CRT (defined as a reduction in LV ESV [>15%] three months after CRT). Thus, TSI seemingly provides information similar to TDI, but it is hoped that TSI will reduce the analysis time, albeit at a step removed from the raw data (28).
The effect of the degree of reverse remodeling after CRT on long-term prognosis was also examined in a study of 141 patients with advanced heart failure. Echocardiography follow-up was performed at three to six months, with clinical follow-up at a mean of 695 ± 491 days. Using an ROC-defined cutoff of 10% reduction in LV ESV, there were 87 responders (61.7%). Responders had significantly lower all-cause mortality (6.9% vs. 30.6%), cardiovascular mortality (2.3% vs. 24.1%), and heart failure events (11.5% vs. 33.3%) than non-responders. Although survival was associated with significantly more dyssynchrony at baseline, in a Cox multivariate regression analysis, the change in LV ESV was the only significant predictor of all-cause and cardiovascular mortality. Left ventricular dyssynchrony at baseline, cause (ischemic vs. non-ischemic), and clinical parameters, including 6-min hallway walk distance and quality of life score, were unable to predict any outcome event, while improvement in New York Heart Association functional class was only marginal (29).
Left atrium.
Doppler parameters provide a beat-to-beat assessment of diastolic function, whereas left atrial volume (LAV) has been suggested to be a marker of the chronicity and severity of diastolic dysfunction (DD). In a retrospective study of 2,042 subjects (30), it was shown that the LAV index progressively increased with worsening diastolic function as assess by conventional Doppler indexes. Both DD and LAV index were predictive of all-cause mortality, but when controlling for DD, LAV index was no longer an independent predictor. In another study (31), LA size was examined in 1,777 competitive athletes. Left atrium size greater than the conventional upper limit of normal (40 mm) was observed in 20% (up to 50 mm in males and 45 mm in females). Multivariate analysis showed that LA enlargement in competitive athletes was largely explained by LV cavity enlargement (R2 = 0.53) and participation in dynamic sports such as rowing and bicycling, but minimally by body size.
Recurrent atrial fibrillation (AF)/stroke.
To determine the relationship of echocardiographic parameters (MR, LA diameter, and LV function) to recurrence of atrial fibrillation and/or stroke, follow-up data from 2,474 patients from the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study who had sinus rhythm resume were analyzed. The cumulative probabilities of at least one AF recurrence/stroke were 46%/1% after 1 year and 84%/5% by the end of the trial (>5 years). Recurrent AF was more likely with larger LA diameters (HR 1.21, 1.16, and 1.32 for mild, moderate, and severe enlargement, respectively). None of these transthoracic echocardiographic measures was associated with risk of stroke. These results confirm several prior studies and suggest that this association between LA size and recurrent AF continues even with the use of current rate-control therapies, including amiodarone. It also helps to clarify prior conflicting results on the role of LA size in predicting recurrent stroke (32).
Mitral regurgitation (MR).
Recent studies have suggested that functional assessment of MR by transesophageal echocardiography (TEE) is a strong determinant of valve reparability and post-operative outcome, with significant incremental value over transthoracic echocardiography (TTE). In a study to determine the relative accuracy of TTE and TEE in predicting the feasibility of valve repair as well as post-operative outcome, results from both techniques for 279 patients who underwent surgery for severe MR in two centers were compared with direct surgical inspection. Agreement with Carpentier’s classification was found in 98% of patients by both techniques. Among 190 patients with mitral valve prolapse or flail mitral valve, the concordance with surgical findings concerning localization of prolapsed or flail segments was 91% for TTE and 93% for TEE (p = 0.4). However, TEE was more accurate for visualizing ruptured chords. The feasibility of repair was significantly influenced by localization of prolapse, with rates of successful repair for isolated P2 lesions, commissural, anterior, and bileaflet prolapse being 99%, 80%, 77%, and 60% (p = 0.001). The feasibility of valve repair was predicted by TTE in 97% of cases, with TEE adding significant information in only two (one patient with bileaflet prolapse, and one with extensive endocarditis and incomplete TTE because of poor image quality). The authors conclude that in experienced hands, TTE can predict repairability without significant incremental value of TEE (33).
Additional reports have: 1) defined criteria for the diagnosis of arrhythmogenic right ventricular dysplasia. Enlargement of the right ventricular outflow tract (RVOT) (generally in the absence of global RV enlargement) was present in 100% of probands with an RVOT diameter of >30 mm in 89% (sensitivity 89% and specificity 86%). The most frequent morphologic abnormality was trabecular derangement (54%), followed by a hyperreflective moderator band in 34% and sacculations in 17% (34); 2) identified a new locus at 13q31.3-q32.1 for the genetic defect in a family mitral valve prolapse (35); 3) expanded our understanding of the pathophysiology of ischemic mitral regurgitation by demonstrating that papillary muscle dysfunction "per se" decreased tethering and MR (36,37); 4) revalidated the importance of vegetation size >10 mm and mobility in predicting embolic complications in infectious endocarditis (38); and 5) identified the multivariate clinical (age >50 years, female gender, pre-operative AF, and concomitant coronary artery bypass grafting) and echocardiographic (LA diameter 46 mm) predictors of overall mortality in patients with hypertrophic obstructive cardiomyopathy after surgical myectomy (39).
|
Summary
|
|---|
While all cardiac imaging modalities continue to improve, echocardiography provides unique clinical information that is readily available at the bedside, in the invasive laboratory, and in the operating room. Recent studies illustrate this potential in heart failure patients where echocardiography can define baseline hemodynamic status, assess the immediate response (and therefore optimize pacing mode) after CRT, and provide unique prognostic information.
|
References
|
|---|
1. Corsi C, Lang RM, Veronesi F, et al. Volumetric quantification of global and regional left ventricular function from real-time three-dimensional echocardiographic images Circulation 2005;112:1161-1170.[Abstract/Free Full Text]2. Kapetanakis S, Kearney MT, Siva A, et al. Real-time three-dimensional echocardiographya novel technique to quantify global left ventricular mechanical dyssynchrony. Circulation 2005;112:992-1000.[Abstract/Free Full Text] 3. Helm RH, Leclerq C, Faris OP, et al. Cardiac dyssynchrony analysis using circumferential versus longitudinal strainimplications for assessing cardiac resynchronization. Circulation 2005;111:2760-2767.[Abstract/Free Full Text] 4. Pemberton J, Li X, Karamlou T, et al. The use of live three-dimensional Doppler echocardiography in the measurement of cardiac outputan in vivo animal study. J Am Coll Cardiol 2005;45:433-438.[Abstract/Free Full Text] 5. Pemberton J, Li X, Kenny A, et al. Real-time 3-dimensional Doppler echocardiography for the assessment of stroke volumean in vivo human study compared with standard 2-dimensional echocardiography. J Am Soc Echocardiogr 2005;18:1030-1036.[CrossRef][Web of Science][Medline] 6. Qin JX, Jones M, Travaglini A, et al. The accuracy of left ventricular mass determined by real-time three-dimensional echocardiography in chronic animal and clinical studiesa comparison with postmortem examination and magnetic resonance imaging. J Am Soc Echocardiogr 2005;18:1037-1043.[CrossRef][Web of Science][Medline] 7. Abdel-Massih T, Dulac Y, Taktak A, et al. Assessment of atrial septal defect size with 3D-transesophageal echocardiographycomparison with balloon method. Echocardiography 2005;22:121-127.[CrossRef][Web of Science][Medline] 8. Bu L, Munns S, Zhang H, et al. Rapid full volume data acquisition by real-time 3-dimensional echocardiography for assessment of left ventricular indexes in childrena validation study compared with magnetic resonance imaging. J Am Soc Echocardiogr 2005;18:299-305.[CrossRef][Web of Science][Medline] 9. Jenkins C, Bricknell K, Marwick TH. Use of real-time three-dimensional echocardiography to measure left atrial volumecomparison with other echocardiographic techniques. J Am Soc Echocardiogr 2005;18:991-997.[CrossRef][Web of Science][Medline] 10. Pemberton J, Li X, Hickey E, et al. Live real-time three-dimensional echocardiography for the visualization of myocardial perfusion—a pilot study in open-chest pigs J Am Soc Echocardiogr 2005;18:956-958.[CrossRef][Web of Science][Medline] 11. Watanabe N, Ogasawara Y, Yamaura Y, et al. Quantitation of mitral valve tenting in ischemic mitral regurgitation by transthoracic real-time three-dimensional echocardiography J Am Coll Cardiol 2005;45:763-769.[Abstract/Free Full Text] 12. Faletra F, La Marchesina U, Bragato R, De Chiara F. Three dimensional transthoracic echocardiography images of tricuspid stenosis Heart 2005;91:499.[Free Full Text] 13. Notomi Y, Lysyansky P, Setser RM, et al. Measurement of ventricular torsion by two-dimensional ultrasound speckle tracking imaging J Am Coll Cardiol 2005;45:2034-2041.[Abstract/Free Full Text] 14. Langeland S, D’hooge J, Wouters PF, et al. Experimental validation of a new ultrasound method for the simultaneous assessment of radial and longitudinal myocardial deformation independent of insonation angle Circulation 2005;112:2157-2162.[Abstract/Free Full Text] 15. Suhling M, Jansen C, Arigovindan M, et al. Multiscale motion mapping: a novel computer vision technique for quantitative, objective echocardiographic motion measurement independent of Doppler: first clinical description and validation Circulation 2004;110:3093-3099.[Abstract/Free Full Text] 16. Wei K, Le E, Jayaweera AR, Bin JP, Goodman NC, Kaul S. Detection of noncritical coronary stenosis at rest without recourse to exercise or pharmacological stress Circulation 2002;105:218-223.[Abstract/Free Full Text] 17. Wei K, Tong KL, Belcik T, et al. Detection of coronary stenoses at rest with myocardial contrast echocardiography Circulation 2005;112:1154-1160.[Abstract/Free Full Text] 18. Grayburn PA. Stress echo without the stressdetection of coronary stenosis at rest by myocardial contrast echocardiography. Circulation 2005;112:1085-1087.[Free Full Text] 19. Senior R, Janardhanan R, Jeetley P, Burden L. Myocardial contrast echocardiography for distinguishing ischemic from nonischemic first-onset acute heart failureinsights into the mechanism of acute heart failure. Circulation 2005;112:1587-1593.[Abstract/Free Full Text] 20. Tong KL, Kaul S, Wang XQ, et al. Myocardial contrast echocardiography versus thrombolysis in myocardial infarction score in patients presenting to the emergency department with chest pain and a nondiagnostic electrocardiogram J Am Coll Cardiol 2005;46:920-927.[Abstract/Free Full Text] 21. Elhendy A, O’Leary EL, Xie F, McGrain AC, Anderson JR, Porter TR. Comparative accuracy of real-time myocardial contrast perfusion imaging and wall motion analysis during dobutamine stress echocardiography for the diagnosis of coronary artery disease J Am Coll Cardiol 2004;44:2185-2191.[Abstract/Free Full Text] 22. Tsutsui JM, Elhendy A, Anderson JR, Xie F, McGrain AC, Porter TR. Prognostic value of dobutamine stress myocardial contrast perfusion echocardiography Circulation 2005;112:1444-1450.[Abstract/Free Full Text] 23. Lyseggen E, Rabben SI, Skulstad H, Urheim S, Risoe C, Smiseth OA. Myocardial acceleration during isovolumic contractionrelationship to contractility. Circulation 2005;111:1362-1369.[Abstract/Free Full Text] 24. Zhang Y, Chan AK, Yu CM, et al. Strain rate imaging differentiates transmural from non-transmural myocardial infarctiona validation study using delayed-enhancement magnetic resonance imaging. J Am Coll Cardiol 2005;46:864-871.[Abstract/Free Full Text] 25. Bruch C, Gotzmann M, Stypmann J, et al. Electrocardiography and Doppler echocardiography for risk stratification in patients with chronic heart failureincremental prognostic value of QRS duration and a restrictive mitral filling pattern. J Am Coll Cardiol 2005;45:1072-1075.[Abstract/Free Full Text] 26. Wang M, Yip G, Yu CM, et al. Independent and incremental prognostic value of early mitral annulus velocity in patients with impaired left ventricular systolic function J Am Coll Cardiol 2005;45:272-277.[Abstract/Free Full Text] 27. Bordachar P, Lafitte S, Reuter S, et al. Echocardiographic parameters of ventricular dyssynchrony validation in patients with heart failure using sequential biventricular pacing J Am Coll Cardiol 2004;44:2157-2165.[Abstract/Free Full Text] 28. Yu CM, Zhang Q, Fung JW, et al. A novel tool to assess systolic asynchrony and identify responders of cardiac resynchronization therapy by tissue synchronization imaging J Am Coll Cardiol 2005;45:677-684.[Abstract/Free Full Text] 29. Yu CM, Bleeker GB, Wing-Hong Fung J, et al. Left ventricular reverse remodeling but not clinical improvement predicts long-term survival after cardiac resynchronization therapy Circulation 2005;112:1580-1586.[Abstract/Free Full Text] 30. Pritchett AM, Mahoney DW, Jacobsen SJ, Rodeheffer RJ, Karon BL, Redfield MM. Diastolic dysfunction and left atrial volumea population-based study. J Am Coll Cardiol 2005;45:87-92.[Abstract/Free Full Text] 31. Pelliccia A, Maron BJ, Di Paolo FM, et al. Prevalence and clinical significance of left atrial remodeling in competitive athletes J Am Coll Cardiol 2005;46:690-696.[Abstract/Free Full Text] 32. Olshansky B, Heller EN, Mitchell LB, et al. Are transthoracic echocardiographic parameters associated with atrial fibrillation recurrence or stroke? Results from the Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) study J Am Coll Cardiol 2005;45:2026-2033.[Abstract/Free Full Text] 33. Monin JL, Dehant P, Roiron C, et al. Functional assessment of mitral regurgitation by transthoracic echocardiography using standardized imaging planes diagnostic accuracy and outcome implications J Am Coll Cardiol 2005;46:302-309.[Abstract/Free Full Text] 34. Yoerger DM, Marcus F, Sherrill D, et al. Echocardiographic findings in patients meeting task force criteria for arrhythmogenic right ventricular dysplasianew insights from the multidisciplinary study of right ventricular dysplasia. J Am Coll Cardiol 2005;45:860-865.[Abstract/Free Full Text] 35. Nesta F, Leyne M, Yosefy C, et al. New locus for autosomal dominant mitral valve prolapse on chromosome 13clinical insights from genetic studies. Circulation 2005;112:2022-2030.[Abstract/Free Full Text] 36. Uemura T, Otsuji Y, Nakashiki K, et al. Papillary muscle dysfunction attenuates ischemic mitral regurgitation in patients with localized basal inferior left ventricular remodelinginsights from tissue Doppler strain imaging. J Am Coll Cardiol 2005;46:113-119.[Abstract/Free Full Text] 37. Popovic ZB, Martin M, Fukamachi K, et al. Mitral annulus size links ventricular dilatation to functional mitral regurgitation J Am Soc Echocardiogr 2005;18:959-963.[CrossRef][Web of Science][Medline] 38. Thuny F, Disalvo G, Belliard O, et al. Risk of embolism and death in infective endocarditis: prognostic value of echocardiography: a prospective multicenter study Circulation 2005;112:69-75.[Abstract/Free Full Text] 39. Woo A, Williams WG, Choi R, et al. Clinical and echocardiographic determinants of long-term survival after surgical myectomy in obstructive hypertrophic cardiomyopathy Circulation 2005;111:2033-2041.[Abstract/Free Full Text]
|
Top
Introduction and validation of...
Refinement and extension of...