The diagnostic role of stress echocardiography in women with coronary artery disease: evidence based review John R. McKeogh

The diagnostic role of stress echocardiography in women with coronary artery disease: evidence based review John R. McKeogh Key points 1) Coronary artery disease in women differs from men in several ways, including their response to many diagnostic tests. 2) Strong comparative data on stress echocardiography for detecting coronary artery disease in women are lacking, but existing data consistently show sensitivity comparable with other high-sensitivity tests and superior specificity, increasing its diagnostic accuracy. 3) Pharmacological stress echocardiography appears to provide superior specificity to exercise stress echocardiography due to difficulties in test execution with the latter that increase the number of false positive results. 4) Current evidence suggests that stress echocardiography should be the first-line diagnostic tool for detecting coronary artery disease in women with chest pain, but other noninvasive imaging tests may be valuable in some patient subgroups. Curr Opin Cardiol 22:429 433. ß 2007 Lippincott Williams & Wilkins. Wolters Kluwer Health Adis, Auckland, New Zealand, an editorial office of Wolters Kluwer Health, Conshohocken, Pennsylvania, USA Correspondence to John R. McKeogh, Adis International Limited, 41 Centorian Drive, Private Bag 65901, Mairangi Bay, Auckland 1311, New Zealand Current Opinion in Cardiology 2007, 22:429 433 Abbreviations 201-SPECT 201-thallium exercise single-photon emission computed tomography ECG electrocardiographic MIBI-SPECT Sestamibi single-photon emission computed tomography STEMI ST-segment elevation myocardial infarction ß 2007 Lippincott Williams & Wilkins 0268-4705 Background A number of significant differences exist between women and men with acute coronary syndromes [1]. Women generally have a more atypical presentation of coronary artery disease. For example, women present more frequently with angina than ST-elevation myocardial infarction (STEMI), and angina is more often uncomplicated than in men, which has led to the perception that angina is usually benign in women [1,2]. Also, compared with men, women present with chest pain on average approximately 10 years later in their life, and initial myocardial infarction occurs approximately 15 20 years later [3]. However, prognosis for women with cardiac manifestations other than angina, e.g. myocardial infarction and coronary artery bypass grafting, is markedly worse than for men (1-year mortality rate in the Framingham study: 45 vs. 10%), and cardiovascular disease is the leading cause of death among women [1,4,5]. Overall, the rate of cardiac events is greater in men, and many of the large scale trials that have led to guidelines for the diagnosis and management of coronary artery disease predominantly involved male patients. The differences in presentation and symptoms combined with the lack of data from female patients mean that diagnosing and managing coronary artery disease in women is often difficult. However, as the prognosis for women with nonanginal disease is poor, an aggressive approach to evaluation of women with chest pain is warranted so coronary artery disease is identified early. It is clear that chest pain is neither a sensitive nor specific indicator for coronary artery disease in women [2]. Exercise electrocardiograph (ECG) testing is valuable for prognostic purposes in women with coronary artery disease, but it has limited diagnostic accuracy, with reported sensitivity and specificity of 61 and 70%, respectively, in one meta-analysis, which are lower than the respective values for men [1,6]. Exercise echocardiography has similar prognostic utility to myocardial perfusion imaging techniques, with both modalities having high negative predictive values for primary and secondary cardiac events [7]. While there are currently inadequate data to determine a difference in the prognostic value of these modalities between men and women, these imaging techniques are valuable for identifying low-risk patients of either gender. In contrast, there appears to be variable diagnostic accuracy of such imaging techniques between women and men. However, 429

430 Imaging and echocardiography imaging techniques have been shown to be more accurate than ECG for diagnosing coronary artery disease in women. The aim of this review is to evaluate the best current evidence for using stress echocardiography, compared with other imaging techniques, to diagnose coronary artery disease in women with chest pain. Procedures reviewed In this review we evaluate exercise or pharmacological (dobutamine or dipyridamole) stress echocardiography. Comparative techniques relevant for this review are ambulatory echocardiography and myocardial perfusion imaging techniques, such as single photon emission computed tomography (SPECT), positron emission tomography and scintigraphy. Search strategy We searched for English language articles in March 2007. We searched for controlled trials using Medline (1966 to date), Embase (1980 to date), the Cochrane Library, and Odyssey (a proprietary database of WK Health). The search terms used were stress echocardiography AND women AND chest pain. Inclusion/exclusion criteria We included prospective randomised and nonrandomised studies investigating stress echocardiography alone or head-to-head with a comparator (a different type of stress echocardiography or a myocardial perfusion imaging technique) in human patients with chest pain; non-randomised studies were included due to the small number of randomised studies identified. The primary focus of this paper is the diagnostic value of stress echocardiography for female patients; however, due to the small number of studies meeting these criteria in female patients exclusively, studies that included both genders with a substantial proportion of female patients (>25%) are also included, but presented separately. Results Three trials evaluating stress echocardiography in exclusively female patients with coronary artery disease were identified (see Table 1) [8 10]. Two of the trials compared pharmacological stress echocardiography with exercise ECG tests and other noninvasive imaging techniques in women who had coronary artery disease diagnosed by coronary angiography [8,9]. The third study compared dobutamine and exercise stress echocardiography with exercise ECG testing in women whose coronary artery disease status was determined by algorithms that included evaluation of the results of the three tests being investigated, and coronary angiography if the results were indeterminate following dobutamine stress echocardiography or if any test returned a strong positive result. In all studies, dobutamine was administered intravenously initially at a low dose, and then incrementally increased to a maximum of 40 mg/kg/min. In the two studies that used coronary angiography solely to diagnose coronary artery disease, specificity of dobutamine echocardiography was found to be superior to both sestamibi SPECT (MIBI-SPECT) and thallium scintigraphy as well as exercise ECG, but sensitivity was not significantly different than any other test evaluated (Table 1) [8,9]. Dobutamine infusion as the stressor also resulted in significantly smaller increases in heart rate and systolic blood pressure compared with exercise in one of the studies [8]. The main limitation of the study by Rollán et al. is that not all participants underwent all three comparator tests [9]. To account for this, dobutamine stress echocardiography data were only included in each comparative analysis for the subgroup of patients who also underwent the applicable comparator test, meaning that different sensitivity, specificity and accuracy results were calculated for dobutamine stress echocardiography for each comparator (Table 1). For diagnostic specificity, all three values and the overall value were 88 89%. Sensitivity values for dobutamine stress echocardiography were more variable (69 80%), but none were significantly different to the comparators. The increased specificity of dobutamine echocardiography compared with exercise ECG and MIBI-SPECT imaging in this study was reflected in significantly greater positive predictive values. There was also a clear trend for increased sensitivity with increasing severity of coronary artery disease. All patients with triple-vessel disease and left main disease were identified with dobutamine echocardiography, while the sensitivities for double-vessel and singlevessel disease were 87% and 55%, respectively. The overall sensitivity was lower than generally observed in male patients, but was probably attributable to a greater number of false negatives associated with single-vessel disease, of which there is a higher incidence in women, compared with multiple-vessel disease. The trend of a slightly greater sensitivity with dobutamine, and specificity with dipyridamole, is consistent with other analyses. The accuracy of both the dobutamine and dipyridamole tests was identical, and also very high. In the study by Laurienzo et al., the semi-invasive transoesophageal echocardiographic technique was used rather than transthoracic echocardiography, which provides lower resolution imaging [8]. The lack of falsepositive results and consequent high specificity associated with echocardiography in this study was attributed at least in part to the use of the transoesophageal approach. The type of echocardiography used in the study by Rollán et al. was not clear [9]. The specificity

Stress echocardiography in women McKeogh 431 Table 1 Summary of studies investigating stress echocardiography alone or head-to-head with a different type of stress echocardiography or a myocardial perfusion imaging technique in women with chest pain Study Participants Challenges Sensitivity (patients with positive test/patients with coronary artery disease) Specificity (1 [patients with negative test/patients without coronary artery disease]) Accuracy ([patients with positive test and coronary artery disease PLUS patients with negative test without coronary artery disease]/ total patients) Laurienzo 1997 [8] Rollan 2002 [9] Sanfilippo 2005 [10] 84 women aged 32 78 years (mean 51) with chest pain who underwent coronary angiography 62 had normal coronary angiographic findings, and 22 had CAD (significant stenosis of 1 major coronary artery) 99 women aged 65 9 years with chest pain who underwent coronary angiography: 57 had normal coronary angiographic findings and 42 had significant CAD (diameter stenosis >50% in a major epicardial artery) 158 women aged 33 79 years (mean 55) with chest pain, no history of cardiac disease and 2 cardiovascular risk factors Transoesophageal dobutamine stress echocardiography Exercise ECG (treadmill test) Thallium scintigraphy (n ¼ 99) Dipyridamole echocardiography (63) Exercise ECG (83) MIBI-SPECT scintigraphy (54) Dobutamine stress echocardiography (n ¼ 47) Exercise echocardiography (57) Exercise ECG (54) ECG: 76% (16/21) Thallium scintigraphy: 86% (19/22) Echocardiography: 82% (18/22) (all participants): 69% (29/42) ECG: 76% (22/29) vs. 72% (21/29) MIBI-SPECT: 88% (23/26) vs. 80% (21/26) 72% (21/29) vs. 79% (23/29) : 75% (3/4) 100% (4/4) ECG: 75% (3/4) ECG: 68% [19/60] Thallium scintigraphy: 80% [12/60] Echocardiography: 100% [0/62] (all participants): 89% (51/57) ECG: 53% (10/19) vs. 89% (17/19) MIBI-SPECT: 57% (16/28) vs. 89% (25/28) 94% (32/34) vs. 88% (30/34) : 97.3% (36/37) 84.8% (38/48) ECG: 93.3% (30/45) ECG: 70% Thallium scintigraphy: 82% Echocardiography: 95% (all participants): 81% ECG: 66 vs. 79% MIBI-SPECT: 72 vs. 85% 84 vs. 84% : 96.3% 84.0% ECG: 67.3% ECG, electrocardiography; HR, heart rate; MIBI-SPECT, Sestamibi single-photon emission computed tomography. Statistically significant difference (between either challenges specified where >1, or specified test (when only one) and all other challenges in the study). Statistical analysis not presented.

432 Imaging and echocardiography of scintigraphy in women tended to be low in these studies, probably due to false positive results associated with other cardiovascular diseases such as mild coronary stenosis, left bundle-branch block and left ventricular hypertrophy. Breast attenuation and smaller left ventricular chamber size in women may also account for lower accuracy associated with scintigraphy. Interpretation of the results of the Sanfillipo study is difficult, as results of statistical analyses detailed in the methods section were not presented with the results [10]. Also, only a few women were ultimately diagnosed with coronary artery disease. Only 4 women diagnosed with coronary artery disease underwent each test, which prevents any rational interpretation of the sensitivity data. Among the 128 coronary artery disease-negative women, there was a trend for greater specificity with echocardiography than with exercise stress ECG. The specificity was also greater with dobutamine stress echocardiography than exercise stress echocardiography, which the study researchers suggest may be due to the greater technical demands associated with performing exercise echocardiography (i.e. rapid position changes for immediate recording of images after exercise has finished) resulting in a greater false-positive rate. The positive predictive value associated with dobutamine echocardiography was 75%, compared with exercise echocardiography and ECG (50 and 36.4%, respectively). It is unfortunate that the statistical significances of these apparent differences were not presented in the study results. The negative predictive values for the three tests were all 97.3%. While it is not possible to draw any conclusions from the data from the Sanfilippo study, they do support the findings of the Laurienzo and Rollán studies with respect to high specificity associated with dobutamine echocardiography. They also suggest that exercise may not be as useful as dobutamine as the stressor for echocardiography, due to the greater difficulties associated with conducting the test, and that where available dobutamine echocardiography should be preferred over exercise echocardiography. When dobutamine echocardiography is not available, a negative exercise echocardiography test is an adequate indicator of absence of coronary artery disease, but a positive test should be confirmed with another test with a good positive predictive value. In addition to the three trials in exclusively female patients presented, a further three trials that included both male and female patients but met all other inclusion criteria were identified [11 13]. None of these studies presented data from female and male patients separately or provided any subanalysis of data stratified by gender. Data from one of these studies was not included, as only one of the 11 subjects was female [13]. In one study involving 45 patients with chest pain of whom 12 were women, dobutamine stress echocardiography and MIBI- SPECT imaging were compared when added to a clinical model of coronary artery disease detection involving patient age and gender, presence of chest pain and dobutamine ECG testing [11]. The addition of MIBI- SPECT improved diagnostic accuracy to a greater extent than the addition of dobutamine stress echocardiography. Overall results showed that sensitivity was significantly greater for MIBI-SPECT than for dobutamine stress echocardiography, but specificity and accuracy were not significantly different. Given the relatively small number of female participants, it is reasonable to assume that any gender differences in the accuracy of these tests could be apparent in these results. The respective sensitivity, specificity and accuracy of 86, 76 and 80% associated with dobutamine stress echocardiography are consistent with values obtained in the studies with exclusively female participants when taking into account the expected lower false-negative results associated with male participants. Similarly, the sensitivity of dobutamine stress echocardiography for detecting coronary artery disease was relatively high at 91% in another study of 26 patients with left bundle branch block, of whom 8 were women [12]. In that study specificity and accuracy of dobutamine stress echocardiography were each 92%. The underlying disease in these patients may have affected these results as the presence of left bundle branch block is known to increase the rate of false positive results. This study also compared dobutamine stress echocardiography with 201-thallium exercise SPECT (201-Tl SPECT) performed using three different approaches. These included a conventional approach (involvement of septal, anterior and/or apical wall), involvement of anterior or septal wall (irrespective of apical wall) and involvement of septum, anterior and apical wall. For the conventional 210-Tl SPECT approach, the specificity of the conventional approach was low and the sensitivity was high. Specificity increased with the third approach to a point where it was no longer significantly lower than for dobutamine stress echocardiography, but the sensitivity decreased to 33% with that approach. It was therefore concluded that dobutamine stress echocardiography is more suitable than 210-Tl SPECT for detecting coronary artery disease in patients with left bundle branch block. Conclusion Although there are a limited number of well-designed studies in the area, the existing data indicate that dobutamine stress echocardiography is associated with superior specificity and similar sensitivity to other tests, and overall high accuracy, for the detection of coronary artery disease in women who present with chest pain. Stress echocardiography may be especially useful for the subpopulation of patients who have a high incidence of false-positive results associated with other noninvasive evaluations of chest pain [8]. Pharmacological stressors

Stress echocardiography in women McKeogh 433 appear to provide more accurate results than exercise, and are also often necessary for women who are not able to exercise at a sufficient intensity to satisfy the requirements of an exercise test. With respect to safety, lower increases in cardiovascular parameters with pharmacological stressors, compared with exercise, must be weighed against the likelihood and impact of drugrelated adverse effects. Dobutamine-associated adverse events reported in the included studies were mild, transient and did not interfere with the tests. While all studies included in this review had limitations, there is some evidence that stress echocardiography should be the first-line diagnostic tool for detecting coronary artery disease in women with chest pain. Scintigraphy may be a good alternative in subgroups of women in whom decreased specificity has been excluded [9]. Further research is needed to enable a definitive recommendation in women with chest pain. Areas that need particular attention include determination of the best stressor, comparisons with other noninvasive imaging techniques, cost effectiveness of the various techniques, and the basis of gender differences associated with diagnostic test results. References 1 Polk DM, Naqvi TZ. Cardiovascular disease in women: sex differences in presentation, risk factors, and evaluation. Current Cardiology Reports 2005; 7:166 172. 2 Gordon EEI. Coronary artery disease in women - the role of diagnostic imaging. Echocardiography 1993; 10:321 330. 3 Lerner DJ, Kannel WB. Patterns of coronary heart disease morbidity and mortality in the sexes: a 26-year follow-up of the Framingham population. Am Heart J 1986; 111:383 390. 4 DaCara JM. Noninvasive cardiac testing in women. J Am Med Womens Assoc 2003; 58:254 263. 5 Kannel WB, Sorlie P, McNamara PM. Prognosis after myocardial infarction: the Framingham Study. Am J Cardiol 1979; 44:53 59. 6 Kwok T, Kim C, Grady D, et al. Meta-analysis of exercise testing to detect coronary artery disease in women. Am J Cardiol 1999; 83:660 666. 7 Metz LD, Beattie M, Hom R, et al. The prognostic value of normal exercise myocardial perfusion imaging and exercise echocardiography: a meta-analysis. J Am Coll Cardiol 2007; 49:227 237. 8 Laurienzo JM, Cannon IR, Quyyumi AA, et al. Improved specificity of transesophageal dobutamine stress echocardiography compared to standard tests for evaluation of coronary artery disease in women presenting with chest pain. Am J Cardiol 1997; 80:1402 1407. 9 Rollán MJ, San Román JA, Vilacosta I, et al. Dobutamine stress echocardiography in the diagnosis of coronary artery disease in women with chest pain: Comparison with different noninvasive tests. Clin Cardiol 2002; 25:559 564. 10 Sanfilippo AJ, Abdollah H, Knott TC, et al. Stress echocardiography in the evaluation of women presenting with chest pain syndrome: A randomized, prospective comparison with electrocardiographic stress testing. Can J Cardiol 2005; 21:405 412. 11 Dibello V, Bellina CR, Gori E, et al. Incremental diagnostic value of dobutamine stress echocardiography and dobutamine scintigraphy (technetium 99m-labeled sestamibi single-photon emission computed tomography) for assessment of presence and extent of coronary artery disease. J Nucl Cardiol 1996; 3:212 222. 12 Tandoğan I, Yetkin E, Yanik A, et al. Comparison of thallium-201 exercise SPECT and dobutamine stress echocardiography for diagnosis of coronary artery disease in patients with left bundle branch block. Int J Card Imaging 2001; 17:339 345. 13 Yip GW, Chandrasekaran K, Miller TD, et al. Feasibility of continuous venous infusion of SonoVue for qualitative assessment of reversible coronary perfusion defects in stress myocardial contrast echocardiography. The International Journal of Cardiovascular Imaging 2003; 19:473 481.