Dobutamine-Atropine Stress Echocardiography in Asymptomatic Healthy Individuals
The Relativity of Stress-Induced Hyperkinesia
Background Interpretation of dobutamine-atropine stress echocardiography (DASE) is based on the assumption that the normal response to dobutamine-atropine infusion is characterized by increased systolic thickening and motion of the left ventricular (LV) walls, whereas a reduction or no change is considered indicative of coronary artery disease. The aim of this study was to quantitatively assess changes in LV dimension and wall motion patterns during DASE in a healthy population.
Methods and Results Forty-two asymptomatic voluntary subjects (22 men) with a mean age of 59 years (range, 31 to 79 years) and a likelihood of <5% for coronary artery disease underwent DASE with digital recording of two-dimensional and M-mode echocardiography at baseline and low-dose and peak infusion rates. Mean end-diastolic and end-systolic LV diameters and areas decreased and wall thicknesses increased progressively throughout the test. Wall motion and thickening increased from baseline to low-dose infusion in nearly all subjects. However, from low-dose to peak infusion, the mean absolute wall motion and relative wall thickening decreased by 13.1% (95% CI, 2.7 to 23.5) and 21.4% (95% CI, 6.4 to 36.4) regardless of age, sex, or use of atropine. Changes in fractional shortening and absolute wall thickening varied considerably, with a decrease observed in 15 and 13 individuals (36% and 31%), respectively.
Conclusions In healthy subjects, measures of wall motion and wall thickening increased from baseline to low-dose infusion but decreased from low-dose to peak infusion. These findings call for revision of the assumptions on which the common analysis of DASE is based.
Dobutamine stress echocardiography (DSE) is gaining popularity as a non–exercise-dependent, noninvasive diagnostic test for detection of coronary artery disease, and in several studies DSE results have correlated well with angiographic1 2 3 4 and scintigraphic5 findings. The protocols of these early studies were based on dobutamine infusion alone. To reach target heart rates in a higher proportion of patients, especially in those treated with β-blockers, a protocol adding atropine to dobutamine infusion is used in many centers.6 7 The results of dobutamine-atropine stress echocardiography (DASE) have also shown a high correlation with angiographic and scintigraphic findings,7 8 9 and the addition of atropine seems to increase the sensitivity of the test with no loss in specificity.7
The cornerstone of stress echocardiography interpretation is the analysis of regional LV wall motion and wall thickening.10 However, the analysis is semiquantitative and subject to variation, and intensive training is required before a maximal diagnostic yield can be expected.11
Precise definitions of normal and abnormal test responses are desirable, not only for routine clinical use but also to facilitate observer training and reproducibility and to form a basis for the development of automatic quantitative wall motion analysis systems. Whereas many investigators have examined changes in LV function during DSE using wall motion analysis, few have tried to characterize these changes by means of quantitative echocardiographic measures,12 13 and to the best of our knowledge, no quantitative data on LV function during DASE have been reported. The interpretation of DASE is based on the assumption that a normal response to graded dobutamine-atropine infusion is increased thickening and motion of the LV walls and that a reduction or no change in wall motion indicates coronary artery disease.7 8 14 As with DSE, however, the wall motion analysis of DASE has so far been performed only semiquantitatively.
The aim of this study was to assess changes in LV dimensions, wall motion, and wall thickening during dobutamine-atropine stress testing in a population with a low probability of coronary artery disease by use of quantitative 2D and M-mode echocardiography.
To determine normalcy rates for noninvasive diagnostic tests, voluntary asymptomatic subjects were selected from the database of the Copenhagen City Heart Study,15 a prospective cardiovascular population study initiated in 1975 to obtain information on risk factors for ischemic cardiovascular disease in the Danish population. The initial sample size consisted of 23 500 men and women ≥20 years of age randomly selected from the Copenhagen population register. From 10 133 respondants in the examination conducted from 1991 through 1994 (third round), an age- and sex-stratified random sample was drawn according to the following criteria: no history of heart disease, no abnormalities at the physical examination or resting ECG, no hypertension, serum cholesterol ≤7 mmol/L, and no medical therapy. Medical history, physical examination, blood biochemistry, blood pressure measurement, and ECG were repeated in 143 respondants and supplemented with an exercise ECG performed with increasing workloads of 25 W every 2 minutes until exhaustion. Secondary exclusion was required in 12 subjects owing to an inability to perform bicycle exercise in 2, inconclusive exercise ECG in 2, ST-segment depression ≥1 mm during exercise ECG in 5, history of angina pectoris in 1, ongoing β-blocker therapy in 1, and an unwillingness to participate in 1. From the remaining 131 voluntary subjects, a random sample of 52 was drawn, all of whom underwent DASE. Only individuals with successful recordings of comparable 2D long-axis views, parasternal or apical, at baseline, low-dose, and peak infusion rates were included in this quantitative study. Furthermore, subjects were excluded whenever obvious echocardiographic angulation errors or insufficient image quality interfered with measurement of any parameter in the 2D long-axis view at any stage. All individuals gave informed consent, and the study was approved by the local ethics committee.
Graded dobutamine infusion was administered through a peripheral arm vein in 3-minute stages at infusion rates of 5, 10, 20, 30, and 40 μg · kg−1 · min−1. When necessary, atropine 0.25 mg was added during peak dobutamine infusion maximally three times at 2-minute intervals in an attempt to increase heart rate to a target of 85% of the age-predicted maximal level (220−age). Reasons for termination of infusion were achievement of target heart rate, maximal drug infusion level at a lower heart rate, development of ventricular tachycardia or ventricular fibrillation, and severe adverse effects, including angina pectoris.
All digital echocardiographic recordings and analyses were performed with a Vingmed CFM 750 ultrasonic scanner (Vingmed Sound) connected to a Macintosh IIci computer equipped with echopac (Vingmed Sound). Two-dimensional echocardiographic recordings were obtained in five standard views: the apical two chamber, apical four chamber, apical or preferably parasternal long axis, parasternal short axis at the chorda level, and parasternal short axis at the midpapillary level. At a frame rate of 25 frames per second, on-line digitalization of cardiac cycles (R to R wave) was carried out at three levels: baseline, low-dose infusion (10 μg · kg−1 · min−1), and peak infusion (up to 40 μg · kg−1 · min−1). An additional acquisition was performed to ensure normalization of the cardiac function (data not included).
M-mode recordings of the left ventricle were guided by the parasternal 2D long-axis and parasternal chorda level short-axis views. On-line digitalization was done at baseline and low-dose and peak infusions. A minimum of five consecutive beats was recorded at each stress level.
LVDd, LVDs, IVSd, IVSs, PWd, and PWs were measured from the 2D long-axis and chorda level short-axis views and from the M-mode recordings. In the short-axis view, additional measurements performed in another two diagonal planes through the center of the cavity described the anterior and inferior free walls and the posteroseptal and lateral free walls. The cavity areas from the parasternal short-axis views at the chorda level were obtained from the end-diastolic and end-systolic frames. In the 2D images, the end-diastolic frame was defined as the frame preceding closure of the mitral valve (ie, the first frame in the cineloop) and the end-systolic frame as the frame with the smallest LV cavity area. Measurements and tracings were carried out according to the principle of the leading edge, including the chorda apparatus and papillary muscles in the cavity area in accordance with the recommendations of the American Society of Echocardiography.16 M-mode measurements also were performed in accordance with these recommendations, except LVDs was determined to be the shortest distance between the walls rather than at the time of peak downward septal motion.17 All measurements were indexed for body surface area. The parameters describing the absolute and relative systolic motion and thickening of the left ventricle were calculated from the following: ΔLVDI=LVDId−LVDIs (mm/m2), FS%=ΔLVDI/LVDId×100 (%), ΔIVSI (ΔPWI)=IVSIs−IVSId (PWIs−PWId) (mm/m2), and ΔIVS% (ΔPW%)=ΔIVSI/IVSId×100 (ΔPWI/PWId×100) (%).
All analyses were performed by one observer blinded to patient identity and clinical characteristics, and the average of three measurements of each parameter was used for statistical analysis. A random sample of 20 studies was reanalyzed by the first observer 3 weeks later to determine the within-observer variability and analyzed by a second observer to assess the between-observer variability of the measurements. Echocardiographic recordings from each stress level were analyzed in separate sessions to blind the observers to results obtained at other levels.
Semiquantitative Wall Motion Analysis
Segmental wall motion analysis was carried out within the frames of a 16-segment model proposed by the American Society of Echocardiography,16 with wall motion scored as follows: 0=hyperkinesia, 1=normokinesia, 2=hypokinesia, 3=akinesia, and 4=dyskinesia, with only systolic endocardial excursion considered. The DASE analysis of digitized cineloops was done by a computer allowing simultaneous, synchronous playback of systole at all stress levels for any 2D echocardiographic view. The analyses were done by a single observer blinded to clinical and quantitative LV dimension and motion data.
The echocardiographic measurements were compared at baseline and low-dose and peak infusions. Differences within and between groups were analyzed with Student’s t test for paired and unpaired data, respectively. ANOVA was used when more than two groups or variables were compared. Correlation analysis was performed by use of Pearson’s r value. Categorical data were analyzed with a χ2 test or Fisher’s exact test whenever appropriate. All probability values given represent two-sided tests. Intraobserver and interobserver variabilities were described as the mean difference between paired observations and the SD of the difference.
Of the 52 screened voluntary subjects, 42 had sufficient 2D long-axis views to allow recording and measurement of all parameters at all stress levels. In this group, the mean age was 59 years (range, 31 to 79 years), and 20 were women. Exercise time and capacity decreased with age as expected. All subjects had negative exercise ECGs, corresponding to an individual likelihood of coronary artery disease <5% as calculated from the tables of Diamond and Forrester,18 including subjects ≥70 years of age in the highest age group of the tables. In an echogenic subgroup of 18 subjects, all 2D long-axis, 2D short-axis, and M-mode measurements were obtained at all stress levels.
In the 42 subjects with sufficient 2D long-axis measurements, peak dobutamine infusion rate was 20, 30, and 40 μg · kg−1 · min−1 in 1, 3, and 38 subjects, respectively. Atropine was added in 26 subjects: 1, 2, and 3 injections of 0.25 mg were given to 9, 9, and 8 individuals, respectively. Infusion was terminated because of achievement of 85% of the age-predicted maximal heart rate in 38 subjects, maximal dose infusion of dobutamine and atropine without reaching this heart rate in 3, and development of nonsustained ventricular tachycardia (24 beats) in 1. Mean heart rate increased significantly, from 63 bpm (range, 44 to 87 bpm) at baseline to 71 bpm (range, 54 to 107 bpm) at low-dose infusion and 139 bpm (range, 113 to 159 bpm) at peak infusion (P≤.05). Mean systolic pressure increased from 126 mm Hg (range, 95 to 182 mm Hg) at baseline to 130 mm Hg (range, 102 to 180 mm Hg) at low-dose infusion (P=NS) and 135 mm Hg (range, 91 to 180 mm Hg) at peak infusion (P≤.05). Peak heart rate and systolic pressure were significantly lower during DASE than exercise ECG (Table 1⇓). Systolic pressure fell below the baseline value in 14 and >10 mm Hg below baseline in 6 subjects without provoking any symptoms. Only minor and transient adverse events occurred: tremor in 5, anxiety in 2, and nausea, dyspnea, and chills in 1 subject each. No subjects complained of chest pain during or after cessation of dobutamine infusion.
Two-dimensional Long-Axis Findings
The rest echocardiograms were considered normal in terms of chamber and valve morphology in all cases. The within- and between-observer variabilities for diastolic measurements given as the mean difference±SD of the difference were 0.2±1.0 and 0.5±1.6 mm for LVDId, 0.6±0.6 and 0.2±0.6 mm for IVSId, and 0.2±0.7 and 0.0±0.7 mm for PWId, respectively. Variations in systolic measurements were of the same order of magnitude. The observer variation was similar at baseline and low-dose infusion. The variation was slightly more pronounced at peak infusion rate: within-observer variations were −0.7±2.2, 0.8±0.8, and 0.4±0.8 mm, and between-observer variations were 0.0±1.5, −0.3±0.7, and 0.0±0.7 mm for LVDId, IVSId, and PWId, respectively. Mean LV dimension indexes at each stress level are given in Table 2⇓. LVDId and LVDIs decreased, and LV wall thickness indexes increased progressively with increasing dobutamine infusion rate. The principal changes in LVDId occurring between the low and peak infusion rates.
Table 3⇓ gives quantitative wall motion parameters. ΔLVDI increased by 38.1% (95% CI, 29.5 to 46.7) from baseline to low-dose infusion but decreased by 13.1% (95% CI, 2.7 to 23.5) from low-dose to peak infusion rate. FS% increased by 40.5% (95% CI, 34.1 to 46.9) from baseline to low-dose infusion and a further 9.8% (95% CI, 1.1 to 18.6) from low-dose to peak infusion. ΔIVSI increased 36.0% (95% CI, 24.6 to 47.4) from baseline to low-dose infusion, but no significant difference was found between low-dose and peak infusion rates. ΔIVS% increased 31.0% (95% CI, 16.7 to 45.2) from baseline to low-dose infusion but decreased 21.4% (95% CI, 6.4 to 36.4), returning almost to the baseline value, at the peak infusion rate. Similar findings were observed in ΔPWI and ΔPW%.
Whereas few individual observations on LV dimensions differed from the general pattern of changes, a substantial variation between subjects was found in motion parameters. Fig 1⇓ shows individual changes in 2D long-axis wall motion parameters as they appear during playback of a cineloop (not indexed). From baseline to low-dose infusion, the LV response was rather uniform, with 71% to 98% of the individuals following the mean trend, but from low-dose to peak rates, decreases in absolute wall motion and wall thickening were seen in 23 (55%) ΔLVDI, 13 (31%) ΔIVSI, and in 13 (31%) ΔPWI of the population. The corresponding numbers for the relative motion and thickening parameters were 15 (36%) FS%, 31 (74%) ΔIVS%, and 27 (64%) ΔPW%. In many individuals, the magnitude of reduction in ΔLVD, ΔIVS%, and ΔPW% from low-dose to peak infusion was comparable to the increase observed from baseline to low-dose infusion (Fig 1⇓). Furthermore, no association was found between the decrease in FS%, ΔIVSI, or ΔPWI and the fall in systolic pressure to below baseline and a fall >10 mm Hg below baseline values. A significant but weak correlation was found between changes in heart rate and changes in LVDId and IVSId from baseline to low-dose infusion (r=.34 and r=.33), and from low-dose to peak infusion (r=−.38 and r=.39; Fig 2⇓). No significant correlations were detected between changes in heart rate or in systolic pressure and changes in any of the remaining dimensions or motion parameters. This was true even for the motion parameters (ΔLVD, ΔIVS%, and ΔPW%) in which a significant reduction in mean value occurred from low-dose to peak infusion (Fig 1⇓).
Comparison of Different Echocardiographic Views
An identical pattern of changes in 2D long-axis LV dimensions and motion indexes during dobutamine-atropine infusion was found in the entire study population (n=42) and the 18 echogenic subjects. Tables 4 and 5 list the results of 2D long-axis, 2D short-axis, and M-mode measurements in the 18 echogenic subjects. For any given parameter, a similar pattern of changes was found regardless of the method of registration and measurement used. However, LV dimension and motion indexes obtained from the 2D long-axis view differed significantly from those obtained with the 2D short-axis or M-mode views. Changes in LV cross-sectional area indexes and absolute and relative systolic area reductions were parallel to the LVDI findings.
Comparison of Different Myocardial Segments
Tables 6 and 7 give the results of measurements in the six basal to midventricular segments visualized at the chorda level short-axis view. Similar patterns of changes during DASE were observed in all three diagonal planes for all parameters investigated. We observed minor but significant differences between diagonal planes in LV cavity and wall dimensions, but they were small compared with changes during DASE. Likewise, small but significant differences in magnitude of wall thickening were found between the six basal to midventricular segments, with thickening in the free wall being greater than in the septal segments at all stress levels.
Significance of Age and Sex
LV dimensions and motion indexes for men and women are delineated in Tables 2⇑ and 3⇑. ANOVA confirmed the presence of significant changes in all dimension and motion parameters during DASE. Measurements of LV dimensions were larger in men (not shown), but women tended to have slightly larger LV dimension and motion indexes at all levels. In addition, the patterns of changes in dimension and motion parameters were similar in subjects below and above 60 years of age.
The decrease in absolute wall motion and relative wall thickening from low-dose to peak infusion rate was independent of sex, age, or the addition of atropine, as illustrated in Fig 4⇓.
Validity of Measurements
The quantitative measurements obtained at one point were considered representative of an entire LV segment in this study. The method is therefore based on the assumption that LV dimensions are close to constant within the segments investigated. This requirement is fulfilled between the tip of the papillary muscles and the mitral valve, corresponding primarily to the border between the basal and the midventricular segments in the 16-segment model commonly used for wall motion analysis. In the apical and extreme basal parts of the left ventricle, however, dimensions may vary considerably, and use of this method in this LV area would not be justified. Because the accuracy of quantitative 2D transthoracic echocardiography depends heavily on the subjects investigated, the operator recording the images, and the observer analyzing them,16 we tried to control for variation and risk of measurement errors. Both the operator and the observers were experienced stress echocardiographers, having performed >500 studies. All measurements were done by one observer, and within- and between-observer variations were within acceptable limits. Patients were selected for echogenicity to provide optimal conditions for echocardiographic recording, and inadequate quality of recordings at any stage in any of the views investigated or obvious angulation errors warranted exclusion from analysis. In a highly selected echogenic subgroup, the 2D measurements were made in two orthogonal planes to control for angulation errors, and M-mode recordings were used as a control for temporal errors occurring during cineloop acquisition or definition of end-diastolic and end-systolic frames. Significant minor differences between measurements obtained with the different methods were expected, and no evidence of timing or angulation errors was found (Tables 4⇓ and 5⇓). Furthermore, the similar pattern of changes identified by the three methods indicates that measurements from 2D long-axis cineloops are valid and representative. In addition, the pattern of changes at different chorda level short-axis planes seems to justify an extension of results form the anteroseptal and posterior walls to the remaining LV walls at the chorda level.
Semiquantitative Wall Motion Analysis
No subject had resting wall motion abnormalities. From baseline to low-dose infusion, an increase in nearly all segments occurred in all subjects, and a decrease was not observed in any segment. From low-dose to peak infusion, a decrease in segmental wall motion was frequent, occurring in all segments, with a reduction from hyperkinesia to a normokinetic state as the most frequent change (Fig 3⇑). In a substantial number of subjects, however, wall motion at peak infusion was reduced to a level below the baseline value, a phenomenon observed most frequently in the basal and midventricular posteroseptal and inferior segments (Fig 3⇑) and rarely in the remaining LV area. This finding was independent of the echocardiographic view chosen. In the basal and midventricular posteroseptal and inferior region, 58 segments had reduced wall motion scores to below the baseline value at peak infusion: by 1 point in 9 (16%), by 2 points in 36 (62%), and by 3 points in 13 (22%) segments.
Interpretation of Results
Dobutamine is a sympathomimetic drug with β-1, β-2, and slight α-1 receptor stimulation properties.19 Continuous intravenous infusion of the drug stimulates cardiac contractility, increases heart rate, and reduces SVR to various degrees, depending primarily on infusion rate.19 Low-dose infusion (≤10 μg · kg−1 · min−1) is generally characterized by increased cardiac contractility, leading to increased stroke volume, while blood pressure is maintained through a reduction in SVR.2 20 During high-dose infusion, stimulation of cardiac contractility persists and heart rate is increased further, whereas the reduction in SVR is less pronounced, resulting in a rise in systemic blood pressure.2 20 The addition of atropine, a cholinergic antagonist, potentiates both the positive chronotropic and the positive inotropic effects of dobutamine21 but prolongs the relative duration of systolic emptying and shortens the diastolic filling phase of the cardiac cycle.22 The progressive decrease in LV chamber diameter and progressive increase in wall thickness during increasing infusion rate demonstrated in the present study are in accordance with these properties of the drug and may be explained by the combined effect of increased contractility and reduced afterload. End-diastolic cavity dimensions were affected only slightly by low-dose dobutamine infusion. The reduction in LV end-diastolic cavity dimensions during high heart rates caused by dobutamine-atropine infusion might be explained by a shortening of the diastole and a relative reduction in venous return. Similar changes in LV cavity dimensions during DSE have been reported in patients without inducible wall motion abnormalities12 or with normal coronary angiograms.13 In addition, previous studies using the thermodilution technique during DSE20 have found stroke volume to be constant from low-dose to peak dobutamine infusion in the absence of detectable myocardial ischemia. The parallel reduction in LV cavity diameter at end systole and end diastole from low-dose to peak infusion in the present study is in accordance with these previous findings. The increased end-diastolic wall thickness at peak infusion is probably a result of the decreased end-diastolic LV cavity size23 and corresponds to findings during rapid atrial pacing24 and DSE.25
The uniform patterns of increase in absolute and relative wall motion and thickening parameters from baseline to low-dose infusion were expected from previous studies of the effects of dobutamine infusion. However, the significant reductions in absolute wall motion and relative wall thickening and the frequent reductions in segmental wall motion scores from low-dose to peak infusion are at variance with previous experience based on semiquantitative wall motion analysis.1 2 3 4 5 6 7 8 9 Our findings of a reduction in end-diastolic cavity size and absolute wall motion are not contradictory to the increase in ejection fraction and constant stroke volume reported in prior studies.20 26 Furthermore, a decreased relative wall thickening may be explained in part by an increased end-diastolic wall thickness at peak infusion. Similar observations have been made in an experimental24 and in a preliminary25 human report. The decreasing FS% from low-dose to peak infusion in some patients was unexpected and suggests a reduction in ejection fraction and stroke volume. A vasovagal reflex mechanism or a functional LV outflow obstruction may be responsible for this phenomenon.27 28
The semiquantitative wall motion data of this study support the frequent reductions in wall motion from low-dose to peak infusion indicated by quantitative determination. This phenomenon was not confined to the basal and midventricular segments, and the most frequent change was a reduction from hyperkinesia to a normokinetic state, similar to the findings from quantitative analysis. In basal and midventricular inferior and posteroseptal segments, the reduction in endocardial excursion tended to be slightly different from that of other segments, as recently reported by Bach et al.29 In the absence of significant coronary disease, they found that dobutamine induced wall motion abnormalities most frequently in the posterior vascular distribution. As they suggested, intermediate-grade coronary stenosis, poor endocardial visualization, and tethering to adjacent fibrous tissue may explain some of their false-positive findings. However, regional intermediate coronary artery disease probably is not responsible for the findings in our unselected asymptomatic population, and even in the study of Bach et al, stress-induced wall motion abnormalities in the basal segments of the posterior distribution were unlikely to have associated coronary lesions.
At baseline, small but significant between-segment differences in measures of wall thickening were observed at the chorda level short-axis view. The finding of this inhomogeneous contraction pattern with a more pronounced wall thickening of the LV free wall compared with the septum is in accordance with previous human studies that used magnetic resonance imaging30 or echocardiography.31 In the present study, these differences in contraction between segments were sustained throughout dobutamine and atropine infusion, confirming the results of Van Rugge et al,30 who used a low-dose infusion protocol. However, one previous echocardiographic study reported a homogeneous LV contraction at peak infusion rate 40 μg · kg−1 · min−1,31 whereas a more pronounced wall thickening in the septal segments during high-dose infusion (30 μg · kg−1 · min−1) was found in another study.25 The use of different quantification methods and the small magnitude of the differences in wall thickening observed between segments may explain these contradictory results. The frequent occurrence of a substantial reduction in endocardial excursion in inferior and posteroseptal segments at peak infusion rate observed in the semiquantitative wall motion analysis, however, does suggest the presence of true heterogenicity in echocardiographic LV wall motion during DASE. Furthermore, apical segments responded different from midventricular segments during low-dose dobutamine infusion in a magnetic resonance imaging study.30
Despite parallel changes in heart rate, blood pressure, LV dimensions, and wall motion during DASE, no close correlations were found between these variables. The inability of changes in these hemodynamics to predict the magnitude of changes in dimensions and motion may be explained by the difference in the dose-response relation demonstrated by Panza et al,31 but a substantial variability between subjects is probably another important cause.
The present quantitative echocardiographic findings indicate that LV wall motion and thickening are significantly different at low-dose and peak infusions and that reduction in motion and thickening from low-dose to peak infusion occurs frequently in a healthy population regardless of age, sex, or use of atropine. This finding challenges the established DASE analysis concept.
Conflict With DASE Analysis Concept
Absolute and relative changes in LV wall motion and thickening form the basis of semiquantitative wall motion analysis.16 In the present study, neither semiquantitative wall motion analysis focused on endocardial excursion nor quantitative wall motion data confirm the DASE analysis concept, although they are in accordance with the expected hemodynamic effects at these drug infusion rates. The DASE concept was originally based on experiences from experimental studies and the established discipline of digital exercise echocardiography. The present study and the study of Perez et al12 demonstrate that the patterns of changes in LV dimensions and wall motion at peak stress during DSE and DASE are quite different from those observed during maximal exercise, whereas they resemble the LV response to rapid atrial pacing.24 32
So, although evidence that DSE and DASE are accurate noninvasive diagnostic tests for coronary disease is accumulating, our results indicate that the specificities of these tests would be very low if interpretation were based solely on wall motion analysis and if a reduction in regional wall motion during increasing drug infusion were considered diagnostic for significant coronary artery disease. Because the presence of latent coronary artery disease in a substantial number of subjects is highly unlikely, one possible explanation of this paradox is that conventional wall motion analysis does not consider LV hyperkinesis, ignoring a possible reduction in the level of hyperkinesia or a change from hyperkinesia to normokinesia. Another explanation could be that other features of regional LV function such as LV dimensions, cavity shape, synchrony of systolic wall motion, or early relaxation are integrated phenomena encompassed in the interpretation of DSE and DASE. The fact that stress echocardiography analysis has proved very difficult to learn, even for echocardiographers with extensive experience in resting wall motion scoring,11 further supports the assumption that stress echocardiography analysis requires the ability to recognize certain complex patterns of changes in regional LV morphology and function that have not been adequately defined.
Dobutamine is often called an exercise simulation agent, but our data cannot confirm this assumption. Neither the hemodynamic response nor the changes in LV dimensions and function during dobutamine stress mimic the responses to exercise; in fact, they resemble the response to rapid atrial pacing.
The finding of frequent stress-induced reductions in segmental LV wall motion and thickening in a healthy population precludes the use of a reduction in wall motion or the absence of hyperkinesia at peak infusion rate as markers of stress-induced myocardial ischemia without a substantial loss of test specificity. Reductions in segmental wall motion below the baseline value, however, were rare in most segments, and segmental wall motion scores below the baseline value in these segments may prove useful for detection of stress-induced myocardial ischemia, provided that this test has an acceptable sensitivity. In the basal and midventricular segments of the posteroseptal and inferior walls, even a severe reduction in wall motion seems to be rather unspecific for myocardial ischemia.
Segmental wall motion analysis obviously has some limitations in its present form with regard to DSE and DASE analyses. Because changes in other indexes of regional LV function such as LV cavity shape, dimensions, wall thicknesses, synchrony of motion, and early relaxation also may be specific markers for inducible myocardial ischemia, the proper variables to be considered in the analysis should be identified, and criteria for normal and abnormal test responses should be defined. Because the LV response varies considerably with the stress modality, such specific definitions are required to enable widespread use of these noninvasive techniques.
This study was funded by grants from the Danish Heart Association and Eli Lilly Denmark A/S.
Selected Abbreviations and Acronyms
|bpm||=||beats per minute|
|DASE||=||dobutamine-atropine stress echocardiography|
|DSE||=||dobutamine stress echocardiography|
|IVSd||=||thickness of the interventricular septum at end diastole|
|IVSId||=||interventricular septum thickness index at end diastole|
|IVSIs||=||interventricular septum thickness index at end systole|
|IVSs||=||thickness of the interventricular septum at end systole|
|ΔIVSI||=||interventricular septum absolute thickening index|
|ΔIVS%||=||interventricular septum relative thickening|
|LVDd||=||LV diameter at end diastole|
|LVDId||=||LV diameter index at end diastole|
|LVDIs||=||LV diameter index at end systole|
|LVDs||=||LV diameter at end systole|
|ΔLVDI||=||LV diameter absolute shortening index|
|PWd||=||posterior wall at end diastole|
|PWs||=||posterior wall at end systole|
|PWId||=||posterior wall thickness index at end diastole|
|PWIs||=||posterior wall thickness index at end systole|
|PWs||=||posterior wall at end systole|
|ΔPWI||=||posterior wall relative thickening index|
|ΔPW%||=||posterior wall absolute thickening|
|SVR||=||systemic vascular resistance|
- Received December 14, 1994.
- Revision received June 20, 1995.
- Accepted July 24, 1995.
- Copyright © 1995 by American Heart Association
Sawada SG, Segar DS, Ryan T, Brown SE, Dohan AM, Williams R, Fineberg NS, Armstrong WF, Feigenbaum H. Echocardiographic detection of coronary artery disease during dobutamine infusion. Circulation. 1991;83:1605-1614.
Marwick T, Willemart B, D’Hondt AM, Baudhuin T, Wijns W, Detry JM, Melin J. Selection of the optimal nonexercise stress for the evaluation of ischemic regional myocardial dysfunction and malperfusion: comparison of dobutamine and adenosine using echocardiography and 99mTc-MIBI single photon emission computed tomography. Circulation. 1993;87:345-354.
Mertes H, Sawada SG, Ryan T, Segar DS, Kovacs R, Foltz J, Feigenbaum H. Symptoms, adverse effects, and complications associated with dobutamine stress echocardiography: experience in 1118 patients. Circulation. 1993;88:15-19.
Baptista J, Arnese M, Roelandt JR, Fioretti P, Keane D, Escaned J, Boersma E, di Mario C, Serruys PW. Quantitative coronary angiography in the estimation of the functional significance of coronary stenosis: correlations with dobutamine-atropine stress test. J Am Coll Cardiol. 1994;23:1434-1439.
Forster T, McNeill AJ, Salustri A, Reijs AE, El Said ES, Roelandt JR, Fioretti PM. Simultaneous dobutamine stress echocardiography and technetium-99m isonitrile single-photon emission computed tomography in patients with suspected coronary artery disease. J Am Coll Cardiol. 1993;21:1591-1596.
Perez JE, Waggoner AD, Davila RV, Cardona H, Miller JG. On-line quantification of ventricular function during dobutamine stress echocardiography. Eur Heart J. 1992;13:1669-1676.
Nyboe J, Jensen G, Appleyard M, Schnohr P. Risk factors for acute myocardial infarction in Copenhagen, part I: hereditary, educational and socioeconomic factors. Eur Heart J. 1989;10:910-916.
Sahn DJ, De Maria A, Kisslo J, Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurement. Circulation. 1978;58:1072-1083.
Pierard LA, Berthe C, Albert A, Carlier J, Kulbertus HE. Haemodynamic alterations during ischaemia induced by dobutamine stress testing. Eur Heart J. 1989;10:783-790.
Landzberg JS, Parker JD, Gauthier DF, Colucci WS. Effects of intracoronary acetylcholine and atropine on basal and dobutamine-stimulated left ventricular contractility. Circulation. 1994;89:164-168.
Hausernova E, Gottdiener JS, Hecht GM, Hausner PF, Weidemann P. Heterogeneity of regional left ventricular wall thickening during dobutamine stress echocardiography in normal subjects. Circulation. 1994;90(suppl I):I-391. Abstract.
Pellikka PA, Oh JK, Bailey KR, Nichols BA, Monahan KH, Tajik AJ. Dynamic intraventricular obstruction during dobutamine stress echocardiography: a new observation. Circulation. 1992;86:1429-1432.
Van Rugge FP, Holman ER, Van Der Wall EE, De Roos A, Van Der Laarse A, Bruschke AVG. Quantitation of global and regional left ventricular function by cine magnetic resonance imaging during dobutamine stress in normal human subjects. Eur Heart J. 1993;14:456-463.
Panza JA, Curiel RV, Laurienzo JM, Unger EF. The normal quantitative myocardial response to dobutamine in humans. Circulation. 1994;90(suppl I):I-391. Abstract.
Pierard LA, Serruys PW, Roeland J, Meltzer RS. Left ventricular function at similar heart rates during tachycardia induced by exercise and atrial pacing: an echocardiographic study. Br Heart J. 1987;57:154-160.