Author + information
- Received August 20, 2002
- Revision received December 11, 2002
- Accepted January 9, 2003
- Published online May 7, 2003.
- Wilson Mathias Jr, MD, PhD, FACC*,* (, )
- Jeane M. Tsutsui, MD, PhD*,
- José L. Andrade, MD, PhD*,
- Ingrid Kowatsch, MD*,
- Pedro A. Lemos, MD*,
- Samira M.B. Leal, MD, PhD*,
- Bijoy K. Khandheria, MD, FACC* and
- José F. Ramires, MD, PhD, FACC*
- ↵*Reprint requests and correspondence:
Dr. Wilson Mathias, Jr., Heart Institute (InCor), University of São Paulo Medical School, Av. Dr. Enéas Carvalho de Aguiar 44, São Paulo, Brazil, CEP:05403-000.
Objectives We studied the value of a rapid beta-blocker injection at peak dobutamine-atropine stress echocardiography (DASE) for the detection of coronary artery disease (CAD).
Background The presence of tachycardia and hyperdynamic wall motion may make it difficult to recognize a new wall motion abnormality (NWMA) at peak stress.
Methods We studied 101 patients (mean age 58.2 ± 9.8 years) who underwent effective DASE and coronary angiography. All patients received a rapid intravenous injection of metoprolol immediately after peak DASE image acquisition. Positivity in combined peak plus post-metoprolol images was defined when there was only peak NWMA, maintenance of peak NWMA, or NWMA detected only after metoprolol injection. Significant CAD was defined as ≥50% stenosis by quantitative angiography.
Results There were 37 patients without and 64 with CAD. The sensitivity, specificity, accuracy, and positive and negative predictive values for the detection of CAD at peak stress were 84%, 92%, 87%, 95%, and 77%, respectively. Five patients with CAD had negative peak images that became positive only after metoprolol. Extension of peak NWMA during metoprolol was observed in 14 patients, and multivessel CAD was detected in 10 of them. The sensitivity, specificity, accuracy, and positive and negative predictive values for peak plus metoprolol images were 92%, 89%, 91%, 94%, and 87%, respectively.
Conclusions The use of metoprolol injected at peak of dobutamine infusion improved the detection of CAD by DASE.
Dobutamine-atropine stress echocardiography (DASE) is a safe and established technique for the diagnosis of significant coronary artery disease (CAD), and beta-blockers have been used to reverse its side effects for many years (1–3). Despite an acceptable accuracy, many factors influence false results and raise questions about the true sensitivity of DASE (4). After the clinical observation of a single case, we hypothesized that one way of overcoming the adverse influences of tachycardia and hyperkinesia was to examine recovery-phase images after a rapid intravenous injection of a selective beta1receptor antagonist at peak DASE in order to improve the detection of a new wall motion abnormality (NWMA).
Between February 2000 and January 2002, DASE was performed in 1,265 patients with known or suspected CAD. Of these, 101 consecutive patients who had an effective DASE study, with beta-blocker injection at peak stress as well as a clinically indicated coronary angiogram within three months, were included in this study. The following exclusion criteria were observed: previous coronary artery bypass graft surgery, acute myocardial infarction, unstable angina, hypertension at rest or during stress (>180 and/or 110 mm Hg), an intervening event between DASE and quantitative coronary angiography (QCA), or contraindications to any drug used in the study. No patient was excluded on the basis of the results of rest echocardiography. The study was approved by our institutional Ethical Committee, and written, informed consent was obtained from all participants.
Dobutamine was administered in a standard fashion, and atropine was injected at the beginning of the stage of 20 μg/kg per min if the heart rate (HR) was <100 beats/min. Immediately after acquisition of peak stress (peak) images, 5 mg metoprolol was injected in 1 min on termination of dobutamine infusion (Fig. 1).
Examinations were performed using an HDI 5000 or Sonos 5500 from Philips Medical Systems, Bothell, Washington. Images were acquired at rest, low dose, and peak stress and after metoprolol, when HR was below 100 beats/min, or at most 3 min after the end of the metoprolol injection. The left ventricle was divided and measured according to the recommendations published elsewhere (5).
Peak images were considered positive when there was inducement of or worsening of a pre-existent wall motion abnormality. Combined peak plus metoprolol images were defined as positive when there were only peak images positivity, maintenance of NWMA observed at peak, extension of peak NWMA, or detection of NWMA only after metoprolol.
Intra- and inter-observer agreements were calculated for a set of 20 stress echocardiograms in a blinded fashion by two independent observers.
QCA was performed with a computer-assisted analysis system, using standard methodology (6), and the images were examined by an experienced angiographer blinded to clinical and echocardiographic data. Significant CAD was defined as ≥50% luminal diameter stenosis in one or more major coronary arteries. Multivessel CAD was defined by the presence of significant left main or two- or three-vessel stenosis.
Continuous data are expressed as the mean ±1 SD, and frequency as proportions. Comparisons between groups for continuous variables were made using the Student tor Mann-Whitney test, when normality was rejected. The chi-square and Fisher exact tests were used for categorical variables. Repetitive measures of variance analysis were performed for comparisons between groups.
Sensitivity, specificity, predictive values, and accuracy were calculated according to standard definitions, and 95% confidence intervals (CIs) were given. The MacNemar test was used for comparisons between DASE and angiographic results.
Multivariate stepwise logistic regression analysis was used to evaluate which variable best-predicted NWMA on metoprolol images. Inter-observer agreement was calculated by the kappa statistic, and intra-observer agreement by Spearman correlation. A two-tailed p value <0.05 was considered statistically significant.
The study population was constituted by 101 patients (58.2 ± 9.8 years old; 61 men). Hypertension was documented in 75 patients, diabetes mellitus in 24 patients, dyslipidemia in 61 patients, cigarette smoking in 27 patients, previous myocardial infarction in 16 patients, and a history of chest pain in 60 patients. Regarding medical therapy, 46 patients were treated with nitrates, 36 with calcium antagonists, 72 with aspirin, and 46 with beta-blockers. Thirty-seven patients had no CAD, and 64 patients had CAD, with multivessel disease detected in 38 (59%) of them.
The dosage of dobutamine used was 30.8 ± 7.6 μg/kg per min. Atropine (0.7 ± 0.5 mg) was used in 91 patients (90%). No major test-related complication was observed.
Detection of CAD by DASE
At peak images, there were 10 false-negative and 3 false-positive results. Analysis of peak images alone presented sensitivity, specificity, and positive and negative predictive values of 84% (95% CI 74% to 91%), 92% (95% CI 78% to 97%), 95% (95% CI 85% to 98%), and 77% (95% CI 63% to 87%), respectively, with an overall accuracy of 87% for detection of CAD.
There were five patients with CAD not detected on peak images in whom NWMAs were observed only after metoprolol injection. Yet, there were increases of sensitivity to 92% (95% CI 83% to 97%), negative predictive value to 87% (95% CI 73% to 94%), and overall accuracy to 91%, with maintenance of specificity and positive predictive values of 89% (95% CI 75% to 96%) and 94% (95% CI 85% to 97%), respectively (Fig. 2). There was only a new patient with normal peak images and marked lateral wall hypokinesia, associated with ST-segment elevation observed only after metoprolol injection (false-positive), who had a 40% diagonal branch lesion.
Patients with single-vessel disease showed an increase in sensitivity from 73% (95% CI 54% to 86%) to 88% (95% CI 66% to 94%). On the other hand, in patients with multivessel disease, the sensitivity at peak stress was 92% (95% CI 79% to 97%), and at peak stress plus metoprolol, it was 95% (95% CI 86% to 98%), with no difference between their values (Fig. 3).
Patterns of response to metoprolol
Negative results of metoprolol injection were obtained in 17 patients (27%). In 12 of them, NWMA was detected at peak DASE with complete recovery after metoprolol, and in five it was negative in both sets of images. On the other hand, 47 patients (73% of all patients with CAD) had positive results with metoprolol (maintenance of NWMA detected at peak stress in 28 patients, extension in 14 patients, and NWMA detected only after beta-blocker injection in 5 patients). Of these five patients, four had single-vessel stenosis with coronary obstructions between 50% and 69% in three, and only one had multivessel disease (Figs. 4 and 5). ⇓Table 1summarizes their hemodynamics.
On multivariate regression analysis, the presence of multivessel disease was found to be the only independent predictor of extension of NWMA (10 of 14 patients) after metoprolol injection (odds ratio [OR] 7.8, 95% CI 2.0 to 30.8; p = 0.0032). No other variable correlated with recovery or maintenance of NWMA (Table 2).
Echocardiographic indexes at rest and during DASE are demonstrated in Table 3.
Intra-observer and inter-observer agreement
We observed excellent correlation between the two measurements for the same observer (intra-observer agreement) for rest (r = 0.97, p < 0.001), low-dose (r = 0.98, p < 0.001), peak (r = 0.96, p < 0.001), and metoprolol images (r = 0.93, p < 0.001). The global inter-observer agreement was 93% (kappa = 0.75, p < 0.001).
Routinely, most echocardiographic laboratories utilize beta-blockers when a significant and hemodynamically important myocardial ischemic event is elicited during DASE (2,3).
The overall accuracy of peak DASE reported in this study resembles that of the current literature for the presence of significant CAD (1). Our results may seem unexpected since a decrease, and not an increase, of sensitivity due to the protective effect of beta-blockade in relieving myocardial ischemia, triggered by a pharmacologic stress test, would be the obvious finding. It has been shown by echocardiography in patients with unstable angina (7)or during balloon angioplasty (8)that beta-blockade ameliorates the electrocardiographic and echocardiographic signs of myocardial ischemia. Also, the use of a beta-blocker is an established therapy in patients with CAD, due to its protective effects on myocardial oxygen consumption, thereby preventing ischemic events. This protective effect is mainly due to a decrease in the rate–pressure product (7–9).
Paradoxically, we demonstrated that the combined peak plus metoprolol images raised the sensitivity, especially in patients with single-vessel CAD, without significant a loss in specificity. To our knowledge, this is the first publication demonstrating the additional diagnostic increment in sensitivity by applying a bolus injection of a beta-adrenergic blocker with the purpose of enabling the observation of NWMA during the early recovery phase.
When injected in the proposed fashion, we demonstrated that there is a decrease in the rate–pressure product, as compared with peak stress, but, importantly, this index was higher at the moment when metoprolol images were acquired than it was at rest. It suggests a sustained oxygen consumption state, prolonging the ischemic effects of dobutamine on the myocardium through the recovery phase. Furthermore, as most of stress-induced ischemia is limited to the subendocardium, ischemia may be masked by the hyperdynamic contractile state from the mid and epicardial layers (10–12). Yet, we observed an elevated rate–pressure product associated with an increase in the left ventricular end-systolic volume index at metoprolol when compared to values at rest. Perhaps this allowed the detection and, sometimes, exacerbation of NWMA that simply was difficult to see at peak stress. Thus, the advantage of using this strategy may be related solely to the enhanced ability to visualize the NWMA.
Another reasonable associated mechanism would be the abrupt reduction of coronary vasodilation through simultaneous blockade of beta1and beta2receptors, by rapid injection of metoprolol, resulting in an abrupt, unopposed alpha receptor–mediated reduction in coronary flow reserve (9,13). A rare and extreme side of this theory would be the occurrence of coronary vasospasm induced by the acute blockage of beta receptors, during high doses of dobutamine, associated with alpha receptor liberation mediated by metoprolol (2). Actually, this could explain the false-positive case we had in a patient that presented with marked lateral wall hypokinesia and electrocardiographic changes only after metoprolol. Therefore, in this setting, myocardial ischemia would result from increased coronary artery vasoconstriction, leading to an increase in coronary resistance and a decrease in coronary artery blood flow (13).
The relatively small number of patients with CAD did not allow a detailed statistical analysis of the relationship between the two groups of different responses to metoprolol and many other important variables. Also, because the study was performed in patients referred to coronary angiography, we recognize that there is a selection bias, but since the patient’s images were compared to themselves, this issue is minimized.
Although an intravenous injection of a standard dose of 5 mg metoprolol does not work similarly for every patient, it has been used for more than a decade in the same fashion and was chosen because we did not want to propose changes in the routine (1–3).
The observation of two specific types of response to metoprolol in patients with CAD suggests that there might be other mechanical, vascular, or metabolic mechanisms involved in this process that we were unable to account for.
This study was designed to document, for the first time, the value of a clinically observed phenomenon; therefore, we recognize that most proposed mechanisms are unknown at this time, warranting further studies.
The most important contribution of this work would be with respect to those patients who achieved >85% of the target HR and still have a negative DASE. In those cases, rapid administration of a beta-blocker at peak DASE will promote an increase in the detection of CAD, especially in patients with single-vessel CAD, a recognized limitation of DASE (4).
This strategy can also be useful in patients whose peak DASE images are doubtful for ischemia, therefore ratifying the NWMA observed at peak stress. On the other hand, in patients with clearly positive peak images, the time to recovery of the induced NWMA is an important predictor of the severity of CAD; therefore, the proposed strategy could compromise this important prognostic information (14).
We sincerely thank Dr. Thomas H. Marwick for thoroughly reviewing this article.
☆ This study was supported by grants from the Heart Institute, University of São Paulo Medical School, São Paulo, Brazil.
- coronary artery disease
- confidence interval
- dobutamine-atropine stress echocardiography
- heart rate
- new wall motion abnormality
- quantitative coronary angiography
- wall motion score index
- Received August 20, 2002.
- Revision received December 11, 2002.
- Accepted January 9, 2003.
- American College of Cardiology Foundation
- Reisenhofer B.,
- Pingitore A.,
- Mathias W.,
- et al.
- Smart C.S.,
- Knickelbine T.,
- Malik F.,
- Sagar K.B.
- Frishman W.,
- Sonnenblick E.H.
- Porter T.,
- Xie F.,
- Silver M.,
- Kricksfeld D.,
- O’Leary E.
- Linka A.Z.,
- Sklenar J.,
- Wei K.,
- Jayaweera A.R.,
- Skyba D.,
- Kaul S.
- Myers J.H.,
- Stirling M.C.,
- Choy M.,
- Buda A.,
- Gallagher K.P.
- Kern M.J.,
- Ganz P.,
- Horowitz J.D.
- Tsoukas A.,
- Ikonomidis I.,
- Cokkinos P.,
- Nihoyanopoulos P.