Comparison Between Dobutamine Echocardiography and Single-Photon Emission Computed Tomography for Interpretive Reproducibility

Comparison Between Dobutamine Echocardiography and Single-Photon Emission Computed Tomography for Interpretive Reproducibility Adele Ferro, MD, PhDa, Teresa Pellegrino, MDb, Letizia Spinelli, MDc, Wanda Acampa, MD, PhDa,d, Mario Petretta, MDc, and Alberto Cuocolo, MDa,d,* Interpretive variability of dobutamine stress echocardiography (DSE) and stress singlephoton emission computed tomography (SPECT) has been previously investigated. However, no study has directly compared the variability of these techniques in the same patient population. We directly compared the interpretive reproducibility of DSE and stress SPECT in patients undergoing both types of pharmacologic stress imaging. Before discharge, simultaneous DSE and SPECT was performed in 56 patients early after a first acute myocardial infarction. Intra- and interobserver concordances were evaluated by exact agreement and ␬ statistic. Intraclass coefficient of correlation was used to assess intra- and interobserver reproducibilities of segmental score analysis. Intraobserver agreement percentages in the identification of patients with ischemia were 98% for SPECT and 91% for DSE (p ⴝ NS) and ␬ values were excellent (>0.80) for both techniques. Interobserver agreement was higher (p <0.01) for SPECT (96%) than for DSE (79%). Similarly, ␬ value was excellent for SPECT (0.92) and only moderate for DSE (0.56). Finally, the intraclass coefficients of correlation for intra- and interobserver reproducibilities were higher for SPECT (0.98 and 0.97, respectively) than for DSE (0.80 and 0.71, respectively; p <0.001 for both). In conclusion, after uncomplicated acute myocardial infarction, stress SPECT imaging has a better interpretive reproducibility than DSE. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;100:1239 –1244)

Dobutamine stress echocardiography (DSE) and cardiac stress single-photon emission computed tomography (SPECT) are widely used for the noninvasive evaluation of patients with ischemic heart disease.1–3 However, the utility of any imaging procedure depends on its interpretive reproducibility, which is 1 of the most important factors in the use of an examination. In this context, observer variability has been investigated with interpretation of DSE4 –7 and SPECT imaging.8 –10 Despite the extensive use of these techniques to detect myocardial ischemia and to stratify risk in patients after acute myocardial infarction, to our knowledge, no systematic comparison of intra- and interobserver variabilities of stress echocardiography and SPECT imaging in the same patient population is currently available. The aim of this study was to directly compare the interpretive reproducibility of DSE and SPECT in the same patients after uncomplicated acute myocardial infarction. Patients underwent simultaneous pharmacologic stress imaging before discharge.

a

Department of Biomorphological and Functional Sciences, University Federico II; bInstitute of Diagnostic and Nuclear Development, SDN Foundation; cDepartment of Clinical Medicine, Cardiovascular and Immunological Sciences, University Federico II; and dInstitute of Biostructure and Bioimages, National Council of Research, Napoli, Italy. Manuscript received March 21, 2007; revised manuscript received May 14, 2007; and accepted May 22, 2007. *Corresponding author: Tel: 39-081-746-2044; fax: 39-081-545-7081. E-mail address: [email protected] (A. Cuocolo). 0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2007.05.041

Methods Patients: We studied 68 consecutive patients with a first uncomplicated acute myocardial infarction referred to the coronary care unit of our university. All patients underwent simultaneous DSE and SPECT imaging with technetium99m sestamibi before discharge (i.e., 3 to 7 days after admission). All patients were treated with systemic thrombolysis during the acute phase and met the following criteria: enzymatic criteria of myocardial infarction, stable hemodynamic conditions, no early postinfarction angina, sinus rhythm, and no conduction disturbances. Patients with history of severe hypertension, valvular heart disease, or other detectable cardiac disorders or severe concomitant illness were excluded. Antianginal medications other than nitrates were discontinued 48 hours before imaging in all patients. For DSE and SPECT studies, the image quality was ranked from 1 (excellent) to 4 (poor) by 2 observers. To exclude the influence of image quality on interpretive reproducibility, cases of poor image quality were excluded. The ethics committee of our university approved the protocol and all patients provided informed consent. Stress test and image acquisition: Dobutamine was infused in 3-minute dose increments of 5, 10, 20, 30, and 40 ␮g/kg body weight per minute with monitoring of symptoms and blood pressure and 12-lead electrocardiography. Test end points were target heart rate (85% of age- and gender-related maximum heart rate), horizontal or downsloping ST-segment depression ⬎2 mm, ST-segment elevation ⬎1 mm, severe angina, systolic blood pressure dewww.AJConline.org

1240

The American Journal of Cardiology (www.AJConline.org)

crease ⬎20 mm Hg, blood pressure ⬎220/120 mm Hg, or clinically important cardiac arrhythmias. If the test end point was not reached at the maximum dose, atropine (ⱕ1 mg) was given. Intravenous ␤-blocker administration was used to reverse the effects of dobutamine/atropine.11 Standard 2-dimensional echocardiography (Sonos 5500; Hewlett-Packard, Andover, Massachusetts) was continuously monitored during the test and a maximum of 10 minutes after the end of the infusion. Technetium-99m sestamibi (740 MBq) was injected intravenously 1 to 2 minutes before completion of dobutamine infusion, and SPECT imaging was performed 60 minutes later. Imaging at rest was performed on a separate day 60 minutes after tracer injection (740 MBq). SPECT acquisition was performed with a multidetector rotating ␥-camera (E.CAM; Siemens Medical Systems, Hoffman Estates, Illinois) equipped with a low-energy, high-resolution collimator and connected to a dedicated computer system. No attenuation or scatter correction was applied. Image analysis: A 17-segment model was used to analyze DSE and SPECT studies, and each segment was assigned to 1 of the major vascular territories.12 On echocardiography, regional wall motion was evaluated on a 4-point scoring system (1, normal; 2, hypokinetic; 3, akinetic; 4, dyskinetic).11 DSE results were classified as 1 of the following: biphasic response (i.e., low-dose improvement followed by a high-dose deterioration), worsening (i.e., direct deterioration), sustained improvement (i.e., low-dose improvement that was maintained at high dose), or no change. Patients with biphasic or worsening response patterns in ⱖ1 segment were considered to have ischemia.11 For SPECT studies, segments were scored on a 4-point system (3, absent tracer uptake; 2, severe reduction; 1, moderate reduction; 0, normal uptake). Scintigraphic variables incorporating the extent and severity of perfusion defects were also calculated.13,14 A summed stress score reflecting ischemia and scar was obtained by adding the scores of the 17 segments of the stress images. A summed score at rest, reflecting the extent of nonreversible defects, was obtained by adding the scores of the 17 segments of the images at rest. Finally, a summed difference score representing the amount of ischemic myocardium was calculated. Patients with scores ⱖ1 were considered to have ischemia. Image interpretation: Image interpretation was performed in separate time-limited sessions to avoid fatigue. Each study was interpreted independently by each reader in a random sequence different from that of any other reader. Intraobserver reproducibility was measured by comparing the images interpreted twice by the same experienced reader (LS and AF for DSE and SPECT, respectively) at an interval of ⱖ6 months. Interobserver reproducibility was measured by comparing the images interpreted by 2 different experienced readers for DSE (LS and MP) and SPECT (AF and TP). Demonstration and maintenance of competence in interpreting stress echocardiograms fulfilled the recommendation of the American College of Cardiology/American Heart Association.1 The observers were blinded to the patient’s name and clinical data and to each other’s results. DSE and SPECT studies were interpreted as ischemic or not

Table 1 Characteristics of study population (n ⫽ 56) Characteristic Age (yrs) Men Systemic hypertension Hypercholesterolemia* Diabetes mellitus Smoker Left ventricular ejection fraction Anterior wall acute myocardial infarction Inferior wall acute myocardial infarction

Value 52 ⫾ 9 48 (86%) 15 (27%) 11 (20%) 9 (16%) 36 (64%) 44 ⫾ 8% 36 (64%) 20 (36%)

* Defined by history or recent cholesterol level at or ⬎90th percentile for age and gender.

Figure 1. Agreement for the classification of patients as nonischemic or ischemic between SPECT study 1 and DSE study 1 (upper panel) and between SPECT study 2 and DSE study 2 (bottom panel). CI ⫽ confidence interval.

ischemic. If ischemic, each area was assigned a coronary vascular territory. Statistical analysis: Continuous variables are expressed as means ⫾ SD and categorical data as percentages. Differences in frequency distribution were compared by Fisher exact test or chi-square test. Intra- and interobserver concordance was expressed as exact agreement and by the ␬ statistic using the following grading system: poor agreement was indicated by a ␬ value ⱕ0.40, moderate agreement was indicated by a ␬ value of 0.41 to 0.60, good agreement was indicated by a ␬ value of 0.61 to 0.80, and excellent agreement was indicated by a ␬ value ⬎0.80.15 A 2-proportion z-test was used to assess difference in exact agreement values. The intraclass coefficient of correlation (ICC) was used to assess intra- and interobserver reproducibilities of segmental score analysis for DSE and SPECT.16 The ICC was calculated as St2 divided by the sum of St2 and S02, in

Coronary Artery Disease/Reproducibility of DSE and SPECT

Figure 2. Intraobserver agreement in the identification of patients with ischemia on stress SPECT and DSE expressed as exact agreement rates (upper panel) and ␬ values (bottom panel).

which S02 is taken to be the within-subject mean square and is a measure of the variance within subjects. St2 is calculated as the between-subject mean square minus the within-subject mean square times K0, in which K0 is the number of replicate measurements, which in this case is 2. Finally, when appropriate, ICC was tested with F being equal to the between-subject mean square divided by the within-subject mean square and tested on n ⫺ 1 and k ⫺ 1 degrees of freedom.16 A p value ⬍0.05 was considered significant. The statistical software used was SPSS 12.0 (SPSS Inc., Chicago, Illinois). Results No patient was excluded for a poor quality SPECT study, but 8 were excluded for poor DSE image quality. Therefore, the final study population comprised 56 patients. The characteristics of the study population are given in Table 1. Myocardial ischemia was detectable in 28 patients (50%) on SPECT (homozonal in 15 and heterozonal in 13) and in 31 patients (55%) on DSE (homozonal in 20 and heterozonal in 11). Agreement between SPECT and DSE in the detection of ischemia is depicted in Figure 1. Intraobserver concordance: For patient analysis, exact agreement percentages were 98% for SPECT and 91% for DSE (p ⫽ NS) and ␬ values were excellent for both techniques (Figure 2). For individual vascular territories, the exact agreement and ␬ values are illustrated in Figure 3. Despite comparable exact agreement, ␬ values were excellent in all territories on SPECT and good to excellent on DSE.

1241

Figure 3. Intraobserver agreement in the identification of ischemia in individual vascular territories on stress SPECT and DSE expressed as exact agreement rates (upper panel) and ␬ values (bottom panel). *p ⬍0.01. LAD ⫽ left anterior descending; LCX ⫽ left circumflex artery; RCA ⫽ right coronary artery.

Interobserver concordance: Exact agreement for patient analysis was higher on SPECT compared with DSE (difference 0.17, 95% confidence interval 0.12 to 0.23, p ⬍0.01; Figure 4). Similarly, the ␬ value was excellent for SPECT (0.92) and only moderate for DSE (0.56). For individual vascular territories, the exact agreement was comparable for SPECT and DSE; however, the ␬ values were excellent in all territories on SPECT and good to excellent on DSE (Figure 5). Reproducibility of segmental score analysis: The intra- and interobserver reproducibilities of segmental score analysis as assessed by the ICC are reported in Table 2. As shown, the intra- and interobserver reproducibilities were greater on SPECT than on DSE. Discussion Interpretive variability is a major methodologic challenge in imaging procedures, and diagnostic methods have been widely examined for intra- and interobserver variabilities because the utility of any technique depends on the accuracy and reproducibility of interpretation. To our knowledge, this is the first study to directly compare the interpretive reproducibility of stress echocardiography and SPECT imaging in the same group of patients undergoing simultaneous dobutamine pharmacologic stress testing after uncomplicated acute myocardial infarction. We used a 17-segment model to analyze DSE and SPECT studies, as this segmentation is more appropriate for the visualization of all segments of the left ventricle and for the assessment of regional

1242

The American Journal of Cardiology (www.AJConline.org)

Figure 4. Interobserver agreement in the identification of patients with ischemia on stress SPECT and DSE expressed as exact agreement rates (upper panel) and ␬ values (bottom panel).

wall motion and myocardial perfusion.12 Our results demonstrate that, after myocardial infarction, stress SPECT has a better interpretive reproducibility than DSE. An important factor affecting the interpretive reproducibility of stress echocardiography is based on image quality. It has been demonstrated that interobserver agreement is excellent (100%) for images with good quality but inadequate (43%) in the presence of poor image quality.4 Therefore, we required high image quality and accordingly excluded 8 patients as a result of poor image quality on echocardiography. Similar to previous investigations,17,18 no patient was excluded from the analysis as a result of poor image quality on SPECT. Previous reports showed that assessment of ischemia by DSE is limited in patients with wall motion abnormalities at rest, such as those with previous myocardial infarction.4 – 6,19,20 Nuclear imaging methods have the intrinsic disadvantages of low spatial resolution and the requirement for radioactive materials.21 However, no significant adverse medical effects have been documented resulting from the use of radioactive agents in standard diagnostic nuclear medicine procedures. Despite these potential limitations, there is an increasing frequency in the application of DSE and SPECT in patients after myocardial infarction.3 Therefore, in our study, we directly compared the interpretive reproducibility of DSE and SPECT in this clinically relevant patient group. It should also be considered that dobutamine-induced changes in regional contractility associated with myocardial ischemia may be subtle and therefore unrecognized or interpreted differently by observers.22,23 As a consequence, high rates of intra- and interobserver disagreement (15% and 18%, respectively) have been found.5 An improved interinstitutional observer interpreta-

Figure 5. Interobserver agreement in the identification of patients with ischemia in individual vascular territories on SPECT and DSE expressed as exact agreement rates (upper panel) and ␬ values (bottom panel). Abbreviations as in Figure 3.

tion agreement on DSE has been reported that involves a second harmonic imaging procedure combined with uniform reading criteria.24 However, there remains significant variability among institutions in stress echocardiographic reading.24 A higher degree of agreement in the interpretation of DSE has been reported for separate observers at the same institution, indicating that implicit agreement on reading criteria could result in improved homogeneity in the interpretation of test results.5,6 The results of the present study show better interobserver interpretive agreement for stress SPECT than for DSE. When individual vascular territories were considered, there was excellent intra- and interobserver concordance on SPECT in all territories. However, on DSE, the interpretive concordance expressed by the ␬ statistic was lower than that of SPECT, with the lowest value in the left circumflex artery territory. These findings are confirmed by the results of segmental score analysis reproducibility assessed by ICC, which is the ratio of the variance of interest over the sum of the variance of interest plus error, providing a measure of reliability.16 Our results show that the ICC is higher on SPECT than on DSE in terms of intra- and interobserver reproducibilities. In this context, it has been suggested that the abnormalities in the inferolateral wall are most likely related to poor endocardial visualization in these regions or could represent regions perceived as hypokinetic.25 Previous studies have suggested that physicians interpreting SPECT images who have met training guidelines might be expected to contribute to good interpretive reproducibility.8,26 In contrast, it is well known that reading DSE is highly

Coronary Artery Disease/Reproducibility of DSE and SPECT

1243

Table 2 Intra- and interobserver reproducibilities of segmental score analysis for dobutamine stress echocardiography (DSE) and stress single-photon emission computed tomography (SPECT) Method

DSE SPECT

Intraobserver Reproducibility

Interobserver Reproducibility

ICC

CI

F*

p Value

ICC

CI

F*

p Value

0.80 0.98

0.77–0.83 0.97–0.99

9.5 55.9

⬍0.001 ⬍0.001

0.71 0.97

0.67–0.75 0.95–0.99

5.2 27

⬍0.001 ⬍0.001

* F represents the between-subject mean square divided by the within-subject mean square and tested on n ⫺ 1 and k ⫺ 1 degrees of freedom. CI ⫽ confidence interval.

subjective and strongly dependent on reader experience.4,9,27,28 Therefore, in the present study, experienced observers interpreted all stress echocardiograms. Reading of stress echocardiograms in clinical practice is still performed without a modality that allows objective quantitative analysis of wall motion response to stress, and the current level of variation of DSE reading is greater than that described for SPECT. In the future, computer algorithms to evaluate wall motion abnormalities may diminish the subjective nature of DSE interpretation and reduce interobserver variability.29 Moreover, new ultrasound imaging methodology developed to enhance detection of contrast agent and to assess perfusion at the tissue level might routinely improve image quality and interpretive reproducibility.21 The present study has some potential limitations that should be considered. Although interpretive variability was greater with DSE, whether this difference is of any clinical value cannot be deduced from this investigation. The study population was relatively small and the stress test was performed with dobutamine, which is not the optimal stress technique for radionuclide imaging. However, this approach represents the usual pharmacologic provocative test in patients with recent myocardial infarction when echocardiography is used. The small number of patients included did not permit us to establish subgroups according to the severity of ischemia. Finally, as we selected only patients with a previous myocardial infarction, the results are not necessarily applicable to unselected patients. 1. Quinones MA, Douglas PS, Foster E, Gorcsan J III, Lewis JF, Pearlman AS, Rychik J, Salcedo EE, Seward JB, Stevenson JG, et al. American College of Cardiology/American Heart Association clinical competence statement on echocardiography: a report of the American College of Cardiology/American Heart Association/American College of Physicians - American Society of Internal Medicine Task Force on Clinical Competence. Circulation 2003;107:1068 –1089. 2. Klocke FJ, Baird MG, Lorell BH, Bateman TM, Messer JV, Berman DS, O’Gara PT, Carabello BA, Russell RO Jr, Cerqueira MD, et al. American College of Cardiology; American Heart Association Task Force on Practice Guidelines; American Society for Nuclear Cardiology. ACC/AHA/ASNC guidelines for the clinical use of cardiac radionuclide imaging— executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2003;108:1404 –1418. 3. Schinkel AF, Bax JJ, Geleijnse ML, Boersma E, Elhendy A, Roelandt JR, Poldermans D. Noninvasive evaluation of ischaemic heart disease: myocardial perfusion imaging or stress echocardiography? Eur Heart J 2003;24:789 – 800. 4. Hoffmann R, Lethen H, Marwick T, Arnese M, Fioretti P, Pingitore A, Picano E, Buck T, Erbel R, Flachskampf FA, Hanrath P. Analysis of interistitutional observer agreement in interpretation of dobutamine stress echocardiograms. J Am Coll Cardiol 1996;27:330 –336. 5. Bjornstad K, Al Amri M, Lingamanaicker J, Oqaili I, Hatle L. Interobserver and intraobserver variation for analysis of left ventricular

6.

7.

8.

9.

10.

11.

12.

13.

14.

15. 16. 17.

18.

19.

wall motion at baseline and during low- and high-dose dobutamine stress echocardiography in patients with high prevalence of wall motion abnormalities at rest. J Am Soc Echocardiogr 1996;9:320 –328. Takeuchi M, Sonoda S, Miura Y, Kuroiwa A. Reproducibility of dobutamine digital stress echocardiography. J Am Soc Echocardiogr 1997;10:344 –351. Picano E, Lattanti F, Orlandini A, Marini C, L’Abbate A. Stress echocardiography and the human factor: the importance of being expert. J Am Coll Cardiol 1991;17:666 – 669. Golub RJ, Ahlberg AW, McClellan JR, Herman SD, Travin MI, Mather JF, Aitken PW, Baron JI, Heller GV. Interpretive reproducibility of stress Tc-99m sestamibi tomographic myocardial perfusion imaging. J Nucl Cardiol 1999;6:257–269. Johansen A, Gaster AL, Veje A, Thayssen P, Haghfelt T, HolundCarlsen PF. Interpretive intra- and interobserver reproducibility of rest/stress 99Tcm-sestamibi myocardial perfusion SPECT in a consecutive group of male patients with stable angina pectoris before and after percutaneous transluminal angioplasty. Nucl Med Commun 2001; 22:531–537. Johansen A, Hoilund-Carlsen PF, Christensen HW, Grupe P, Veje A, Vach W, Haghfelt T. Observer variability in the evaluation of dualisotope Tl-201/Tc-99m sestamibi rest/stress myocardial perfusion SPECT in men and women with known or suspected stable angina pectoris. J Nucl Cardiol 2004;11:710 –718. Armstrong WF, Pellikka PA, Ryan T, Crouse L, Zoghbi WA. Stress echocardiography recommendations for performance and interpretation of stress echocardiography. Stress Echocardiography Task Force of the Nomenclature and Standards Committee of the American Society of Echocardiography. J Am Soc Echocardiogr 1998;11:97–104. Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, Pennell DJ, Rumberger JA, Ryan T, Verani MS. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart. A Statement of Healthcare Professional from the Cardiac Imaging Committee of the Council on Clinical cardiology of the American Heart Association. Circulation 2002;105:539 –542. Acampa W, Spinelli L, Petretta M, De Lauro F, Ibello F, Cuocolo A. Prognostic value of myocardial ischemia in patients with uncomplicated acute myocardial infarction: direct comparison of stress echocardiography and myocardial perfusion imaging. J Nucl Med 2005;46: 417– 423. Hachamovitch R, Berman DS, Kiat H, Cohen I, Cabico JA, Friedman J, Diamond GA. Exercise myocardial perfusion SPECT in patients with known coronary artery disease. Circulation 1996;93:905–914. Fleiss JL. Statistical methods for rates and proportions. 2nd ed. New York: Wesley and Sons; 1981:217–225. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979;86:420 – 428. 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. Senior R, Lahiri A. Enhanced detection of myocardial ischemia by stress dobutamine echocardiography utilizing the “biphasic” response of wall thickening during low and high dose dobutamine infusion. J Am Coll Cardiol 1995;26:26 –32. Kvitting JP, Wigstrom L, Strotmann JM, Sutherland GR. How accurate is visual assessment of synchronicity in myocardial motion? An in vitro study with computer-simulated regional delay in myocardial

1244

20.

21. 22.

23.

24.

The American Journal of Cardiology (www.AJConline.org)

motion: clinical implications for rest and stress echocardiography studies. J Am Soc Echocardiogr 1999;12:698 –705. Armstrong WF, O’Donnell J, Ryan T, Feigenbaum H. Effect of prior myocardial infarction and extent and location of coronary disease on accuracy of exercise echocardiography. J Am Coll Cardiol 1987;10: 531–538. Cuocolo A, Acampa W, Imbriaco M, De Luca N, Iovino GL, Salvatore M. The many ways to myocardial perfusion imaging. Q J Nucl Med Mol Imaging 2005;49:4 –18. Borges AC, Pingitore A, Cordovil A, Sicari R, Baumann G, Picano E. Heterogeneity of left ventricular regional wall thickening following dobutamine infusion in normal human subjects. Eur Heart J 1995;16: 1726 –1730. Carstensen S, Ali SM, Stensgaard-Hansen FV, Toft J, Haunso S, Kelbaek H, Saunamaki K. Dobutamine-atropine stress echocardiography in asymptomatic healthy individuals. Circulation 1995;92:3453– 3463. Hoffmann R, Marwick TH, Poldermans D, Lethen H, Ciani R, van der Meer P, Tries HP, Gianfagna P, Fioretti P, Bax JJ, et al. Refinements in stress echocardiographic techniques improve inter-institutional

25.

26.

27. 28. 29.

agreement in interpretation of dobutamine stress echocardiograms. Eur Heart J 2002;23:821– 829. Bach DS, Muller D, Bernard JG, Armstrong W. False positive dobutamine stress echocardiograpms:characterization of clinical, echocardiographic and angiographic findings. J Am Coll Cardiol 1994;24: 928 –933. Danias PG, Ahlberg AW, Travin MI, Mahr NC, Abreu JE, Marini D, Mann A, Mather JF, Boden WE, Heller GV. Visual assessment of left ventricular perfusion and function with electrocardiography-gated SPECT has high intraobserver and interobserver reproducibility among experienced nuclear cardiologists and cardiology trainees. J Nucl Cardiol 2002;9:263–270. Hoffmann R, Hanrath P. The scourge of subjective interpretation. Eur Heart J 1995;16:1459 –1461. Varga A, Picano E, Dodi C, Barbieri A, Pratali L, Gaddi O. Madness and method in stress echo reading. Eur Heart J 1999;20: 1271–1275. Zwirn G, Beeri R, Gilon D, Akselrod S. Automatic endocardialboundary detection in low mechanical-index contrast echocardiography. IEEE Trans Biomed Eng 2006;53:2310 –2322.