Postsystolic shortening is a sensitive maker of myocardial ischemia. The aim of this study was to investigate whether diastolic dyssynchrony imaging is useful for the objective interpretation of dobutamine stress echocardiography.
Postsystolic shortening was detected by using tissue Doppler imaging displacement timing analysis: the delays of the displacement peaks from end-systole were displayed from green to red, depending on the preset time window on diastolic dyssynchrony imaging. Dobutamine stress echocardiography was performed in 59 patients with suspected coronary artery disease who presented with normal left ventricular wall motion at rest (age range, 44–83 years; 20 women). The optimal time windows for diastolic dyssynchrony imaging at rest and at peak dobutamine were determined by receiver operating characteristic analysis by measuring the delays of the displacement peaks in the left ventricular myocardial segments. Diastolic dyssynchrony imaging was performed using time windows of 100 msec at rest and 80 msec at peak dobutamine. The diagnostic power of diastolic dyssynchrony imaging was assessed with quantitative coronary angiography as the gold standard (>50% diameter stenosis) both at rest and at peak dobutamine.
Coronary artery disease was present in 37 patients (63%). Diastolic dyssynchrony imaging at peak dobutamine predicted the presence of coronary artery disease with sensitivity of 89%, specificity of 77%, predictive accuracy of 85%, positive predictive value of 79%, and negative predictive value of 81%. Diastolic dyssynchrony imaging at rest yielded sensitivity of 62%, specificity of 73%, predictive accuracy of 66%, positive predictive value of 79%, and negative predictive value of 53%. Importantly, diastolic dyssynchrony imaging demonstrated excellent intraindividual (97%) and interindividual (90%) agreement.
Diastolic dyssynchrony imaging is useful in the objective interpretation of dobutamine stress echocardiography.
Although dobutamine stress echocardiography is an established method for detecting coronary artery disease in clinical practice, the interpretation of dobutamine stress echocardiography requires expertise. Numerous efforts have been made to date to quantify wall motion for the purpose of making objective interpretation, for example, velocity measurements by tissue Doppler imaging (TDI), strain and strain rate by TDI, displacement by TDI, and, more recently, those by two-dimensional pixel tracking. Although these quantification techniques appear promising, applying them in clinical practice has not yet gained popularity, because of reproducibility issues as well as cumbersome postprocessing procedures.
Postsystolic shortening (PSS) is a delayed ejection motion of the myocardium occurring after aortic valve closure during a generally prolonged isovolumic relaxation time, which is closely associated with myocardial ischemia. PSS was detected on dobutamine stress echocardiography when myocardial ischemia was induced. Thus, if PSS can easily be visualized during dobutamine stress echocardiography, objective interpretation should be possible.
We previously developed TDI software that readily visualizes the presence of PSS on two-dimensional echocardiography, referred to as detection of diastolic abnormality on dyssynchrony imaging. This diastolic dyssynchrony imaging uses tissue Doppler–derived displacement timing analysis, which is independent of the amplitude of the data, and is consequently much less dependent on the Doppler angle. In addition, because it includes neither difference nor differentiation calculation, and because it does not measure amplitude, diastolic dyssynchrony imaging has demonstrated excellent intraindividual and interindividual reproducibility. Thus, the purpose of this study was to investigate whether diastolic dyssynchrony imaging is useful in making objective interpretations of dobutamine stress echocardiography.
We prospectively enrolled 62 consecutive patients with suspected coronary artery disease between May 2006 and July 2008, who were referred for diagnostic dobutamine stress echocardiography at the Kansai Rosai Hospital Cardiovascular Center, met the inclusion and exclusion criteria of this study, and agreed to participate. Consent was obtained to perform coronary angiography in addition to echocardiography, including TDI. Patients presenting with abnormal echocardiographic results at rest, such as wall motion abnormalities, significant valvular diseases, dilated or restrictive cardiomyopathy, left ventricular hypertrophy (interventricular septal or posterior wall thickness ≥ 12 mm), and pulmonary hypertension (tricuspid valve regurgitation velocity ≥ 2.5 m/sec), were excluded. Patients with histories of myocardial infarction, previous coronary angioplasty or bypass grafting, atrial fibrillation or flutter, pacemaker implantation, left bundle branch block, or congestive heart failure were also excluded. After enrollment, three patients were further excluded from the analysis because of inadequate ultrasound images. The remaining 59 patients (mean age, 67 years; range, 44–83 years; 20 women) were included in the study. The study protocol was approved by the ethics committee of Kansai Rosai Hospital. All patients gave written informed consent.
All patients underwent a standard dobutamine stress echocardiographic protocol, as described earlier. Dobutamine was given in 3-min increments from 10 to 40 μg/kg/min, and up to 2 mg of atropine was given, as needed, to achieve 85% of age-predicted maximum heart rate. Criteria for terminating the test included completion of the protocol, severe ischemia evidenced by extensive new wall motion abnormalities, severe angina or ST-segment changes, systolic blood pressure > 240 or < 100 mm Hg, serious ventricular arrhythmia, patient intolerance, and serious side effects caused by dobutamine.
Echocardiography and TDI
All patients underwent routine echocardiography before the study to exclude echocardiographic abnormalities at rest. Standard echocardiography and color-coded TDI were performed using a cardiac ultrasound diagnostic apparatus (Aplio SSA-770A; Toshiba, Tokyo, Japan) with a 3.6-MHz sector transducer. Echocardiograms were obtained in the left lateral position at end-expiration. TDI was performed in the standard apical planes, including four-chamber, two-chamber, and long-axis views. Tissue Doppler images were digitally recorded both at rest and at peak dobutamine.
Classic Wall Motion Analysis by Two-Dimensional Echocardiography
Left ventricular wall motion was assessed by an expert cardiologist (Y.H.) with >10 years of experience in dobutamine stress echocardiography who was blinded to patients’ clinical and angiographic data. A regional wall motion score was obtained for each segment of the standardized 16-segment model. Regional myocardial performance was classified as normal, mildly hypokinetic, severely hypokinetic, and akinetic or dyskinetic. Ischemia was identified by new or worsening wall motion abnormalities with stress.
Diastolic Dyssynchrony Imaging
Diastolic dyssynchrony imaging was performed as previously described using the digitally stored TDI data obtained both at rest and at peak dobutamine. In brief, the velocity measured by TDI was integrated over time using a TDI tracking technique, and the timing of the displacement peak was calculated for each pixel. End-systole was automatically determined by the software in the identical beat using the Doppler velocity signal as being at the timing of the sum of the Doppler velocities from the whole plane to be zero. The delay of the displacement peak from the end-systole was color coded from green (no delay) to red (delay greater than the time window selected). In this way, diastolic dyssynchrony imaging color coded the delay of the displacement peak from end-systole occurring in the whole left ventricle and thereby displayed the anatomic distribution of PSS as a color-coded parametric image on the two-dimensional echocardiogram. Separate time windows were used for diastolic dyssynchrony imaging at rest and for that at peak dobutamine, on the basis of the results of receiver operating characteristic analysis described in the following section. Coronary artery disease was judged present by diastolic dyssynchrony imaging when the part of the left ventricle was segmentally color coded red. Patchy patterns were excluded because they were considered to be affected by noise. Intraobserver agreement of dobutamine stress diastolic dyssynchrony imaging was assessed by a single investigator (T.O.) in 30 randomly selected patients on two separate occasions. Interobserver agreement was also assessed in the same patient population by two independent observers (T.O. and S. Ohata).
Determination of the Time Window at Rest and at Peak Dobutamine
Before diastolic dyssynchrony imaging, we obtained the displacement curves from TDI velocity data using online software (TDI-Q; Toshiba) at the annular and mid left ventricular levels in three standard apical planes (a total of 12 points). Myocardial displacement along the ultrasound beam direction was calculated by temporal integration of the velocity at each myocardial point on the beam. Importantly, this software is capable of tissue Doppler tracking along the beam direction, thereby enabling the accurate measurement of displacement. End-systole was estimated from the Doppler velocity in the identical beat, as described in the previous section. The maximum delay from end-systole to the peak displacement curve (Δ T ) was calculated for each patient ( Figure 1 ). Subsequently, the optimal cutoff value of Δ T for discriminating normal subjects from those with coronary artery disease was determined by receiver operating characteristic analysis at rest (98.5 msec) and at peak dobutamine (75.0 msec). On the basis of the receiver operating characteristic analysis, a time window of 100 msec was used at rest and that of 80 msec at peak dobutamine for diastolic dyssynchrony imaging.
Coronary angiography was performed using a standard technique within 3 weeks of dobutamine stress echocardiography. Stenosis severity was measured by quantitative coronary angiography using an automated edge detection system (CASS; Pie Medical Imaging BV, Maastricht, The Netherlands). A maximal luminal diameter stenosis of >50% in any plane was defined as significant. Quantitative coronary angiography was done by a single expert cardiologist (T. Ishihara), who was blinded to the clinical and echocardiographic data other than angiography.
Numerical variables are expressed as mean ± SD. Unpaired t tests were used for comparisons between patients with and without coronary artery disease. We considered results significant for p values < .05. The optimal cutoff values of the duration of the time window set from end-systole for discriminating normal subjects from patients with angiographic coronary artery disease were determined by the receiver operating characteristic curves at rest and at peak dobutamine.
Coronary artery disease was diagnosed by quantitative coronary angiography in 37 patients (63%) among 59 patients enrolled in this study. Twenty-eight patients (47%) had single-vessel disease, and nine patients (15%) had multivessel disease. The left anterior descending coronary artery (LAD) was stenosed in 26 patients (44%, LAD lesions); non-LAD lesions (the left circumflex artery [ n = 7] and/or the right coronary artery [ n = 6] were stenosed without the involvement of the LAD) were observed in 11 patients (19%). No significant differences were found between patients with and without coronary artery disease regarding age, risk factors, systolic and diastolic blood pressure, heart rates, β-blocker use, and calcium antagonist use. The proportion of men and nitrate use were greater in patients with coronary artery disease than in those without coronary artery disease ( Table 1 ).
|Patients with coronary artery disease||Patients without coronary artery disease|
|Variables||( n = 37)||( n = 22)||P|
|Age (years)||63 ± 12||65 ± 9||NS|
|Coronary risk factors|
|Rest blood pressure (mm Hg)||137/67||135/72||NS/NS|
|Peak dobutamine blood pressure (mm Hg)||155/70||143/66||NS/NS|
|Rest heart rate (beats/min)||62 ± 10||64 ± 9||NS|
|Peak dobutamine heart rate (beats/min)||122 ± 18||126 ± 16||NS|