- Email: [email protected]
Correlation of myocardial doppler velocity response to exercise with independent evidence of myocardial ischemia by dual-isotope single-photon emission computed tomography
Correlation of Myocardial Doppler Velocity Response to Exercise With Independent Evidence of Myocardial Ischemia by Dual-Isotope Single-Photon Emission Computed Tomography Agne`s Pasquet,
MD,
Guy Armstrong, MD, Curtis Rimmerman, Thomas H. Marwick, MD, PhD
MD,
and
Myocardial Doppler velocity (MDV) imaging may provide an objective correlate of ischemia, thereby reducing the expertise needed for interpreting stress echocardiography and improving its reproducibility. This study sought to independently validate the results of exercise MDV imaging with single-photon emission computed tomography (SPECT) perfusion imaging in 116 patients (age 60 ⴞ 12 years, 28 women) referred for exercise SPECT for diagnostic or prognostic assessment of coronary artery disease. Two-dimensional echocardiography was performed with simultaneous color MDV data acquisition before and after exercise treadmill testing. MDV data were processed off-line to display myocardial velocity profiles in each segment at rest and peak exercise. SPECT was analyzed using a 16-segment model and segments were classified as normal or showing resting or stress defects. Resting defects within segments showing normal function were attributed to attenuation. Color MDV data were compared with SPECT results, and a multivariate analysis (including exercise and SPECT results) was performed to identify the determinants of
the exercise MDV response. Patients exercised maximally (peak rate-pressure product 27.6 ⴞ 6.1 ⴛ 103), and SPECT was abnormal in 33 patients. Of the 1,333 left ventricular segments evaluable by SPECT and MDV, 1,217 segments were classified as normal, 43 showed a stress defect, and 73 a rest defect. Segmental comparison of thallium findings and MDV showed that segments with a rest defect had a lower velocity at rest and stress than normal segments (p <0.001). Segments with a stress defect had a marked reduction in peak exercise velocity and less increment in velocity than normal segments. Heart rate, functional capacity, and presence of abnormally perfused segments were independent predictors of myocardial velocity at peak exercise. Thus, color MDV correlates with independent evidence of ischemia, although it is also influenced by exercise capacity and left ventricular function. This technique may permit a feasible approach to quantitation of exercise echocardiography. 䊚2000 by Excerpta Medica, Inc. (Am J Cardiol 2000;85:536 –542)
uring the past decade, exercise echocardiography has evolved as a cost-effective alternative to nuD clear techniques for detecting and managing coronary
experimental protocol.7 In patients, MDV has also been used to quantify changes during dobutamine echocardiography,8 –12 and may be used to recognize ischemic segments.11 However, MDV has only been compared with results of wall motion scoring,11 and has not been compared with other techniques used for the diagnosis of coronary disease. Moreover, the use of exercise rather than dobutamine stress permits the assessment of exercise capacity, and constitutes a more physiologic stress, but the integration of MDV with exercise echocardiography is not well defined. The aims of this study were to validate the use of MDV during exercise echocardiography by comparison with single-photon emission computed tomography (SPECT) and to analyze the determinants of peak MDV at exercise.
disease with comparable accuracy.1,2 However, subjectivity in the interpretation of wall motion abnormalities3 and the need for adequate training4 are frequently cited as limitations to the use of this modality. Several attempts have been made to quantify regional function during stress echocardiography.5,6 The recent development of tissue Doppler measurement of myocardial Doppler velocity (MDV) has introduced a parameter that is relatively independent of two-dimensional (2-D) echo image quality, which has been used to quantify changes in regional function during an From the Department of Cardiology, Cleveland Clinic Foundation, Cleveland, Ohio; and the University of Queensland, Brisbane, Australia. This study was supported in part by the Clive and Vera Ramaciotti Foundation, Sydney, Australia. Manuscript received July 21, 1999; revised manuscript received and accepted October 7, 1999. Address for reprints: Thomas H. Marwick, MD, University Department of Medicine, Princess Alexandra Hospital, DH2W, Ipswich Road, Brisbane, Queensland 4012, Australia. E-mail: tmarwick@ medicine.pa.uq.edu.au.
536
©2000 by Excerpta Medica, Inc. All rights reserved. The American Journal of Cardiology Vol. 85 March 1, 2000
METHODS
Study patients: We studied 116 patients (age 60 ⫾ 12 years, 88 men) undergoing exercise SPECT for evaluation of known or suspected coronary artery disease. Patient agreement to participate in the study and availability of personnel and equipment were the 0002-9149/00/$–see front matter PII S0002-9149(99)00807-3
only selection criteria. Exercise testing was performed in 32 patients for diagnosis of coronary disease and for prognostic assessment (including follow-up of known disease) in 84 patients. The pretest probability of disease, derived from age, gender, and symptoms was 41 ⫾ 25%.13 A history of myocardial infarction was obtained in 38 patients, and 75 had previous revascularization. Most had multiple risk factors for coronary artery disease: 69 had systemic hypertension, 77 had hypercholesterolemia (defined by total cholesterol ⬎5.5 mmol/L, low-density lipoprotein cholesterol ⬎3.5 mmol/L, or treatment with lipid-lowering therapy), and 21 were diabetic. -Blocking agents were given to 45 patients, and 27 were taking calcium antagonists. Exercise dual-isotope SPECT: Rest thallium SPECT scintigraphy was usually performed in the morning preceding the stress test to identify the presence of resting defects. Images were obtained using a standard protocol immediately after injection of 1 mCi of thallium. We used a double- or triple-headed camera with a high-resolution collimator in a circular 360° orbit. All subjects exercised on a treadmill using Bruce or modified protocols selected in accordance with their age and exercise capacity. Before exercise a medical history was recorded and an intravenous line was secured. The electrocardiogram was monitored continuously during the test for ST changes and arrhythmias, and a 12-lead electrocardiogram was recorded before starting exercise, at the end of each stage, and after exercise. Blood pressure was measured at each stage and during the recovery period. At peak exercise, 20 mCi of technetium sestamibi was injected intravenously and the patient was asked to exercise for 1 more minute. Tests were symptom-limited, with the usual end points being fatigue, dyspnea, severe angina or severe ST-segment changes, decrease or exaggerated increase in blood pressure, or serious arrhythmia.14 Peak exercise images were obtained 1 hour after the end of the exercise using the same acquisition protocol as the resting images. Transaxial images were reconstructed using a Hamming filter in the short axis and horizontal and vertical long axes in a side-by-side rest-stress display. Images were interpreted using the same 16-segment model used for evaluation of MDV, and efforts were made during rotation and selection of imaging planes to reproduce the echo image planes based on “landmarks” at the junction of the right ventricular free wall and septum, the membranous septum, apex, and base. Segments were classified as normal, rest, or stress perfusion defects based on the comparison of rest and stress perfusion scan by observers blinded to clinical and exercise echocardiography data using a previously validated quantitative color scale.15 Ischemia by SPECT was defined as a stress perfusion defect (change of tracer activity by ⬎15%) in an area with normal perfusion at rest. A perfusion defect present on rest and stress perfusion scan (⬎20% less than maximum) was considered to be scar, unless resting function was normal, in which case the perfusion defect
was ascribed to attenuation and considered to be normal. Exercise echocardiography with color myocardial Doppler: Color MDV was recorded using commer-
cially available equipment (System FiVe, Vingmed Sound AB, Horten, Norway) with a 2.5-MHz transducer using standard settings for tissue Doppler.16,17 Images were obtained in long- and short-axis parasternal and apical long-axis views, 4- and 2-chamber views at rest, and as quickly as possible after completion of the treadmill test. Color MDV was acquired in real time, superimposed on 2-D images using a sector angle of 60° and a frame rate of at least 72 frames/s; the Nyquist limit was set at 19 cm/s and increased if necessary to avoid aliasing. Acquisition of MDV data did not require additional time compared with traditional 2-D echo. A complete cardiac cycle was acquired in each view at rest and immediately after exercise, and stored in digital format for off-line analysis as well as being recorded on VHS videotape. Wall motion analysis was performed by experienced observers blinded to other data. Regions showing a new or worse wall motion abnormality after exercise were identified as ischemic. Segments with abnormal wall motion at rest, without worsening after stress, were identified as infarcted. All other segments were identified as normal. MDV analysis was performed on color MDV cine loops derived from polar data. A region of interest was positioned in the middle of apical, mid- and basal segments of each wall,18 and a myocardial velocity profile was displayed throughout the cardiac cycle in each location (Figure 1). Analysis of the systolic component of the velocity profiles permitted measurement of peak systolic velocity at rest and after exercise in 12 defined segments. Statistical analysis: Results are reported as mean ⫾ SD. Continuous and discrete variables were compared using t tests and chi-square tests, respectively. Because neither SPECT nor wall motion scoring are completely accurate as indexes of disease, we defined normal segments based on normalcy with both techniques. Segmental SPECT interpretation was compared with systolic MDV by analysis of variance single-factor analysis with Bonferroni correction. A p value ⬍0.05 was considered statistically significant. A multiple linear regression model was used to assess the relative contribution of exercise variables and ischemia or scar to postexercise MDV.
RESULTS
Exercise responses: Clinical and exercise data are summarized in Table I. The mean ejection fraction was 64 ⫾ 13%. At exercise, 98 patients (85%) reached ⬎85% of the age-predicted heart rate. Twenty-seven patients complained of chest discomfort during the test; ischemic ST-segment changes occurred in 20 patients and 18 had ischemia by 2-D echocardiography. Feasibility and reproducibility: Of the 1,392 segments available, MDV was interpretable in 96%. Of the 59 segments without interpretable MDV data, 32 (54%) were located in the anterior wall. Based on the
CORONARY ARTERY DISEASE/MYOCARDIAL DOPPLER VELOCITY AND EXERCISE SPECT
537
FIGURE 1. Myocardial velocity profile through the cardiac cycle in a normal segment at rest (left) and during exercise (right). Velocity profile displays systolic velocity (S) and early (E) and late (A) diastolic components. At stress, both diastolic waves are frequently fused (right).
FIGURE 2. Color MDV 2-D echo of 2-chamber view (end-systolic freeze frames) with corresponding myocardial velocity profile in an inferior ischemic segment at rest (left, arrow shows site of ischemia) and during exercise (right). After stress, the increase in MDV is blunted (right arrow corresponds to peak systolic velocity).
reanalysis of 70 segments (rest or exercise), correlation between 2 readings of peak systolic MDV measurements was 0.92. The mean time to process the color MDV loops and interpret the velocity profiles for the 12 segments at rest and stress was 38 ⫾ 8 minutes/patient. 538 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 85
Correlation with thallium SPECT: Patients were divided into those with normal rest and stress SPECT (n ⫽ 83) and those with abnormal SPECT at rest (n ⫽ 14), at stress (n ⫽ 8), or both (n ⫽ 11). The mean left ventricular ejection fraction in the normal group was 58 ⫾ 11% and in the abnormal group 63 ⫾ 14% (p MARCH 1, 2000
TABLE I Clinical Data and Hemodynamic Parameters in Patients With and Without Normal SPECT
Age (yrs) Systemic hypertension Diabetes mellitus Hypercholesterolemia‡ Ejection fraction (%) Pretest probability of coronary artery disease (%) Complete test (⬎85% predicted peak heart rate) Incomplete test (⬍85% predicted peak heart rate) General fatigue Shortness of breath Chest pain/ST abnormalities Medical therapy  blocker Calcium channel antagonist Rest heart rate (beats/min) Systolic blood pressure rest (mm Hg) Diastolic blood pressure rest (mm Hg) Rest heart pressure product (⫻ 1,000) Peak heart rate (beats/min) Systolic blood pressure peak (mm Hg) Diastolic blood pressure peak (mm Hg) Peak rate pressure product (⫻ 1,000)
Normal (n ⫽ 83)
Abnormal (n ⫽ 33)
58 ⫾ 11 46 12 55 67 ⫾ 12 39 ⫾ 24
63 ⫾ 14* 23 9 22 58 ⫾ 15† 47 ⫾ 26
74
24*
9
9
8 1 0
8 0 1
31 20 71 ⫾ 12 140 ⫾ 20
14 7 71 ⫾ 12 142 ⫾ 22
85 ⫾ 11
85 ⫾ 12
9.85 ⫾ 2.1 150 ⫾ 20 190 ⫾ 26 91 ⫾ 14
10.20 ⫾ 2.7 141 ⫾ 17* 175 ⫾ 25*
TABLE II Change in Systolic Myocardial Velocities Occurring During Exercise in Normal Segments
Rest cMDV (cm/s) Exercise cMDV (cm/s) ⌬ cMDV %⌬ cMDV
Basal (n ⫽ 185)
Mid (n ⫽ 183)
Apex (n ⫽ 113)
⫾ ⫾ ⫾ ⫾
⫾ ⫾ ⫾ ⫾
⫾ ⫾ ⫾ ⫾
5.6 10.6 4.9 91
1.3* 3.9 3.2* 7.8 2.8* 3.9 62 115
1.3* 2.2 3.0* 4.5 2.9* 2.4 116 148
1.7* 3.4* 3.7* 416
*p ⬍0.001. Results are expressed as mean ⫾ SEM. Basal ⫽ basal segments; cMDV ⫽ color myocardial Doppler velocity; Mid ⫽ midsegment; ⌬ ⫽ difference between myocardial velocity at rest and exercise.
TABLE III Comparison of Rest and Stress MDV According to SPECT Results (normal, stress defect, or rest defect) Normal (n ⫽ 1,217) Rest cMDV (cm/s) Exercise cMDV (cm/s) ⌬ cMDV (cm/s) %⌬ cMDV
4.4 8.2 3.8 99
⫾ ⫾ ⫾ ⫾
1.6 3.4 2.9 106
Stress Defect (n ⫽ 43) 4.4 6.5 2.3 76
⫾ ⫾ ⫾ ⫾
Rest Defect (n ⫽ 73)
1.7 3.5 ⫾ 1.7*‡ 3.1† 5.1 ⫾ 2.8* 2.9† 1.6 ⫾ 2.1* 109 61 ⫾ 92§
*p ⬍0.0001 versus normal; †p ⬍0.005 versus normal; ‡p ⬍0.05 versus stress defect; §p ⬍0.01 versus normal. Results are expressed as mean ⫾ SEM. Abbreviations as in Table II.
89 ⫾ 14
28.48 ⫾ 5.89 24.77 ⫾ 5.71*
*p ⬍0.05; †p ⬍0.001. ‡ Defined by total cholesterol ⬎5.5 mmol/L, low-density lipoprotein (LDL) cholesterol ⬎3.5 mmol/L, or lipid-lowering therapy.
⬍0.005). Among the 83 patients with normal rest and stress SPECT, 67 had also normal exercise echocardiography based on wall motion scoring (concordance 81%). Thirty patients were abnormal on exercise echo and SPECT and 3 patients showed only abnormal SPECT (concordance 91%). SPECT analysis in segments with color MDV measurements identified 1,217 as normal, 73 showed a resting defect compatible with scar, and 43 showed an exercise-induced perfusion defect compatible with ischemia. Among the 1,217 segments with normal rest and stress SPECT scan, 1,136 had normal wall motion on exercise echocardiography (concordance 93%). Among 116 segments showing abnormal perfusion pattern on stress SPECT, 91 had wall motion abnormalities at peak stress (concordance 78%). NORMAL SEGMENTS: In 1,217 normal SPECT segments, myocardial velocity increased by 99 ⫾ 106% at peak exercise. Based on analysis of 31 patients with normal exercise echocardiography and SPECT, without a history of coronary artery disease, resting velocity in basal segments was significantly greater than in mid- or apical segments (5.6 ⫾ 1.3 vs 3.9 ⫾ 1.3 cm/s, and 2.2 ⫾ 1.7 cm/s; p ⬍0.001). This gradient remained with exercise (10.6 ⫾ 3.2 vs 7.8 ⫾ 3.0 cm/s, and 4.5 ⫾ 3.4 cm/s; p ⬍0.001) (Table II). The velocity
increment with stress was similar in basal and midsegments (p ⫽ NS). ABNORMAL SEGMENTS: Segments with rest defects had a lower MDV than normal segments (p ⬍0.0001) at rest and stress (Table III, Figure 2). Ischemic segments (stress-induced defect) were not different from normal segments at rest but had a lower MDV at peak exercise (6.5 ⫾ 3.1 vs 8.2 ⫾ 3.4 cm/s, p ⬍0.05) (Figure 3). Segments with scar had a significantly lower MDV at rest than ischemic segments (3.5 ⫾ 1.7 vs 4.4 ⫾ 1.7 cm/s; p ⬍0.05), but peak MDV was not different (p ⫽ 0.07). The normal velocity increment with exercise was 3.8 ⫾ 2.9 cm/s in normal segments. This increment was significantly greater than the velocity increment for ischemic segments (2.3 ⫾ 2.9 cm/s, p ⬍0.005) and scar segments (1.6 ⫾ 0.3 cm/s, p ⬍0.0001). The increment in myocardial velocity was comparable for scar and ischemic segments (2.3 ⫾ 2.9 vs 1.6 ⫾ 2.1 cm/s, p ⫽ NS). The increment of MDV expressed as a proportion of resting MDV showed a wide overlap in this derived value but only rest defects differed significantly from normal segments (61 ⫾ 92% vs 99 ⫾ 106%, p ⬍0.05). Correlates of the MDV response to exercise. In univariate analysis, peak exercise MDV correlated with clinical (age, risk factors, cardiac history), exercise testing (workload, hemodynamic responses), echocardiographic (segment location and overall left ventricular function), and perfusion defect extent (p ⬍0.01). However, in a multiple linear regression model, only perfusion defect extent, exercise capacity, peak heart rate, and segment location were independent predictors of peak MDV.
CORONARY ARTERY DISEASE/MYOCARDIAL DOPPLER VELOCITY AND EXERCISE SPECT
539
FIGURE 3. Color MDV 2-D echo of 2-chamber view (end-systolic freeze frames) with corresponding myocardial velocity profile in an inferior scar segment at rest (left, arrow shows site of scar) and during exercise (right). Peak systolic velocity (right arrow) is subnormal and shows a small increment with exercise.
DISCUSSION These results show that measurement of MDV is feasible during exercise echocardiography and correlates with results of exercise SPECT. Resting MDV in segments with resting perfusion defects is significantly lower than normal. The increment of MDV induced with exercise is significantly blunted in ischemic or scar segments. Maximal heart rate and functional capacity, as well as the presence of coronary artery disease, are the major determinants of peak exercise myocardial velocities. Quantitative approaches to stress echocardiography:
Several techniques, including centerline methods,19 color kinesis,5 and automatic boundary detection6 have been proposed to help interpretation of stress echocardiography. Nevertheless, none have proved to be “user friendly,” and despite limitations of wall motion scoring, the qualitative approach remains most widely used. MDV has the benefit of being quantitative and relatively independent of 2-D image quality. MDV can be displayed in spectral patterns20,21 or color coded.16,17,22 Because heart rate decreases rapidly after exercise, we chose to use color MDV to permit rapid acquisition of all data in the postexercise period. The software used for this study allowed us to circumvent previous limitations in the use of color Doppler: the temporal resolution is improved by using a high frame-rate (⬎70 frames/s), and off-line postprocessing is easy, although still time-intensive. MDV response to exercise: Conventional interpretation of stress echocardiography focuses on endocardial excursion or wall thickening in a radial direction. The apical MDV method examines base-to-apex motion, engendered by the piston-like action of the base toward the apex, which stays quite immobile during 540 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 85
the cardiac cycle.23,24 This interrogation of longitudinal rather than radial contraction is parallel with the ultrasound beam, implying that the angle dependence of Doppler is not a major limitation to accuracy. Indeed, previous stress MDV studies,9 –11,25,26 including our own,12,27 using the apical window have shown that the velocity in the basal and midzones corresponds closely to the qualitative assessment of regional function in the radial direction. Nonetheless, this correlation may merely reflect radial thickening due to longitudinal shortening, so that correlation with an independent marker of ischemia is an important development. In experimental settings, MDV has been shown to accurately reflect changes in regional and global left ventricular function induced with dobutamine, esmolol, or ischemia.7,25 In humans, induced ischemia during percutaneous transluminal coronary angioplasty balloon inflation is associated with a marked decrease in MDV and followed by recovery after balloon deflation.8 Likewise, the velocity of normal segments increases in response to dobutamine infusion,26 or exercise,27 and ischemic segments have a lower velocity response.11,12 From the apical views, the subnormal velocity increment of ischemic segments to exercise may reflect the susceptibility of longitudinal subendocardial fibers to ischemia. Comparison of echo and scintigraphy: Nuclear techniques are frequently considered as an optimal test for diagnosis of coronary artery disease because of wide experience, and because interpretation is based on objective criteria and quantification. Compared with scintigraphy, stress echo is a more recently introduced technique and current interpretation is based on subjective criteria. Although studies comparing both techMARCH 1, 2000
niques have shown similar overall accuracy,1,2 some discrepancies between the tests are inevitable, because they assess different aspects of the physiology of coronary artery disease: stress echo considers regional function and SPECT evaluates relative perfusion. For example, some segments that appear akinetic or severely hypokinetic due to necrosis affecting the subendocardial layers have a low velocity at rest, but have normal perfusion on SPECT due to the surrounding area of normal tissue. Moreover, SPECT is not a “gold standard” for the diagnosis of coronary artery disease. Segments with resting perfusion defects may show normal metabolism using positron emission tomography.28 Artifacts can play an important role in classification of segments and may lead to overestimation of the presence of scar. For this reason, perfusion data were integrated with resting wall motion, and apparent “fixed defects” in the setting of normal function were considered to be due to attenuation, and classified as normal. Study limitations: The apex only demonstrates slight motion during the cardiac cycle, so apical velocities are very low and show high variability. Moreover, the complex myocardial architecture of the heart influences the MDV that can be developed in different segments, and a large population (including a normal group) is needed to designate normal and abnormal ranges in each segment. The regional variability of MDV and problems with assessing the apex are important limitations of this technique, and we would therefore propose MDV to be used more as an adjunct to help interpretation of exercise echocardiography rather than a substitute for subjective interpretation. MDV measures velocity relative to the transducer, including velocity due to myocardial thickening, translation of the heart, and motion due to breathing. The latter can be minimized by taking images during apnea. The relative contribution of translation is unknown, although the presence of measurable velocities during ischemia or in the presence of scar suggests that this is a significant influence on MDV. Thus, the velocity increment with stress is colored by tethering of adjacent segments. The effects of translation and tethering may be reduced or removed by the use of myocardial velocity gradient rather than MDV.29 The gradient technique has been used in anterior and posterior segments in the parasternal view; we did not use this approach because of our plan to compare longaxis function in the apical view. Diastolic function is known to be altered at an early stage by ischemia. Unfortunately, the design of this exercise study only permitted data acquisition after exercise, at which time tachycardia inhibits the ability to separate E and A waves. In our previous experience,12 the feasibility of analyzing diastolic parameters was ⬍30%. However, analysis of MDV at submaximal heart rates may make regional diastolic assessment more feasible. This study is based on matching the MDV and SPECT results in the same myocardial segments, and the design is dependent on being sure that adequate matching between SPECT and echo segments has
occurred. We have developed considerable experience with this process, using landmarks to match the location of individual segments. However, it is expected that some discrepancies between the techniques arise from a mismatching of the segments. 1. Quinones MA, Verani MS, Haichin RM, Mahmarian JJ, Suarez J, Zoghbi WA.
Exercise echocardiography versus 201Tl single-photon emission computed tomography in evaluation of coronary artery disease. Analysis of 292 patients. Circulation 1992;85:1026 –1031. 2. Fleischmann KE, Hunink MG, Kuntz KM, Douglas PS. Exercise echocardiography or exercise SPECT imaging? A meta-analysis of diagnostic test performance, JAMA 1998;280:913–920. 3. Hoffman R, Lethen H, Marwick T, Arnese M, Fioretti P, Pingitore A, Picano E, Buck T, Erbel R, Flachskampf FA, Hanrath P. Analysis of interinstitutional observer agrement in interpretation of dobutamine stress echocardiograms. J Am Coll Cardiol 1996;27:330 –336. 4. Picano E, Lattanzi F, Orlandini A, Marini C, L’Abbate A. Stress echocardiography and the human factor: the importance being expert. J Am Coll Cardiol 1991;17:666 – 669. 5. Lang RM, Vignon P, Weinert L,. Echocardiographic quantification of regional left ventricular wall motion with color kinesis. Circulation 1996;93:1877–1885. 6. Perez JE Waggoner AD, Davila Roman VG, Cardona H, Miller JG. On line quantification of ventricular function during dobutamine stress echocardiography. Eur Heart J 1992;13:1669 –1676. 7. Gorcsan J III, Strum DP, Mandarino WA, Gulati VK, Pinsky MR. Quantitative assessment of alterations in regional left ventricular contractility with color-coded tissue Doppler echocardiography. Comparison with sonomicrometry and pressure-volume relations. Circulation 1997;95:2423–2433. 8. Bach DS, Armstrong WF, Donovan CL, Muller DWM. Quantitative Doppler tissue imaging for assessment of myocardial velocities during transient ischemia and reperfusion. Am Heart J 1996;132:721–725. 9. Fontanet HL Pueblo JA, Davis MG, Lockeby M, Thannikary L, Schoken DD. Quantitative dobutamine stress echocardiography utilizing Doppler tissue imaging. J Am Coll Cardiol 1996;(suppl):64A: 10. Von Bibra H, Tuchnintz A, Firscke C, Schu¨hlen H, Scho¨ming A. Doppler tissue imaging of the left ventricular myocardium-initial results during pharmacologic stress. J Am Coll Cardiol 1995;(suppl):57A: 11. Katz WE, Gulati VK, Mahler CM, Gorcsan J III. Quantitative evaluation of the segmental left ventricular response to dobutamine stress by tissue Doppler echocardiography. Am J Cardiol 1997;79:1036 –1042. 12. Yamada E, Garcia M, Thomas JD, Marwick TH. Myocardial Doppler velocity imaging—a quantitative technique for interpretation of dobutamine echocardiography. Am J Cardiol 1998;82:806 – 809. 13. Diamond GA, Forester J. Analysis of probability as an aid in the clinical diagnosis of coronary artery disease. N Engl J Med 1979;300:1350 –1358. 14. Ellestad M, ed. Stress testing protocol. In: Stress Testing. Philadelphia: FA Davis, 1986:157–186. 15. Go RT, Marwick TH, MacIntyre WJ, Saha GB, Neumann DR, Underwood DA, Simpfendorfer CC. A prospective comparison of rubidium-82 PET and thallium-201 SPECT myocardial perfusion imaging utilizing a single dipyridamole stress in the diagnosis of coronary artery disease. J Nucl Med 1990;12:1899 – 1905 16. Sutherland GR, Stewart MJ, Groundstroem KW, Moran CM, Fleming A, Guell-Peris FJ, Riemersma RA, Fenn LN, Fox KA, McDicken WN. Color Doppler myocardial imaging: a new technique for the assessment of myocardial function. J Am Soc Echo 1994;7:441– 458. 17. Wilkenshoff UM, Sovany A, Wigstrom L, Olstad B, Lindstrom L, Engvall J, Janerot-Sjoberg B, Wranne B, Hatle L, Sutherland GR. Regional mean systolic myocardial velocity estimation by real-time color Doppler myocardial imaging: a new technique for quantifying regional systolic function. J Am Soc Echo 1998; 11:683– 692. 18. Schiller NB, Shah PM, Crawford M, Demaria A, Devereux R, Feigenbaum H, Gutgesell H, Reichek N, Sahn D, Schnittger I. Recommendations for quantification of the left ventricle by two dimensional echocardiography. American Society of Echo Committee on Standards, Subcommittee on Quantitation of Two Dimensional Echocardiograms. J Am Soc Echo 1989;2:258 –267. 19. Sheehan FH, Bolson EL, Dodge HT, Mathey DG, Schofer J, Woo HW. Advantages and applications of the centerline method for characterizing regional ventricular function. Circulation 1986;74:293–305. 20. Garcia MJ, Rodriguez L, Ares M, Griffin BP, Klein AL, Stewart WJ, Thomas JD. Myocardial wall velocity assessment by pulsed Doppler tissue imaging: characteristic findings in normal subjects. Am Heart Journal 1996;132:648 – 656. 21. Hada Y, Itoh N, Tohyo Y, Yonekura K, Tamiya E, Kiritani H. Intramyocardial pulsed Doppler echocardiography as a new modality for evaluation of left ventricular wall motion: assessment in normal subjects. J Cardiol 1996; 28:85–92. 22. Donovan CL, Armstrong WF, Bach DS. Quantitative Doppler tissue imaging
CORONARY ARTERY DISEASE/MYOCARDIAL DOPPLER VELOCITY AND EXERCISE SPECT
541
of the left ventricular myocardium: validation in normal subjects. Am Heart J 1995;130:100 –104. 23. Jones CJ, Raposo L, Gibson DJ. Functional importance of the long axis dynamic of the human left ventricle. Br Heart J 1990;63:215–220. 24. Adam M, Hoglund C, Thorstand C. Longitudinal systolic shortening of the left ventricle an echocardiographic study in subjects with and without preserved global function. Clin Physiol 1992;12:443– 452. 25. Derumeaux G, Ovize M, Loufoua J, Andre-Fouet X, Minaire Y, Cribier A, Letac B. Doppler tissue imaging quantitates regional wall motion during myocardial ischemia and reperfusion. Circulation 1998;97:1970 –1977. 26. Gorcsan J III, Deswal A, Mankad S, Mandarino WA, Mahler CM, Yamazaki N, Katz WE. Quantification of the myocardial response to low-dose dobutamine
542 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 85
using tissue Doppler echocardiographic measures of velocity and velocity gradient. Am J Cardiol 1998;81:615– 623. 27. Pasquet A, Armstrong G, Beachler L, Lauer M, Marwick TH. Analysis of segmental myocardial Doppler velocity as a quantitative adjunct to exercise echocardiography. J Am Soc Echo 1999;12:901–912. 28. Brunken R, Schwaiger M, Grover-McKay M, Phelps ME, Tillisch J, Schelbert H. Positron emission tomography detect tissue metabolic activity in myocardial segments with persistent thallium perfusion defects. J Am Coll Cardiol 1987;10:557–567. 29. Tsutsui H, Uematsu M, Shimizu H, Yamagishi M, Tanaka N, Matsuda H, Miyatake K. Comparative usefulness of myocardial velocity gradient in detecting ischemic myocardium by a dobutamine challenge. J Am Coll Cardiol 1998;31: 89 –93.
MARCH 1, 2000
MD,
Guy Armstrong, MD, Curtis Rimmerman, Thomas H. Marwick, MD, PhD
MD,
and
Myocardial Doppler velocity (MDV) imaging may provide an objective correlate of ischemia, thereby reducing the expertise needed for interpreting stress echocardiography and improving its reproducibility. This study sought to independently validate the results of exercise MDV imaging with single-photon emission computed tomography (SPECT) perfusion imaging in 116 patients (age 60 ⴞ 12 years, 28 women) referred for exercise SPECT for diagnostic or prognostic assessment of coronary artery disease. Two-dimensional echocardiography was performed with simultaneous color MDV data acquisition before and after exercise treadmill testing. MDV data were processed off-line to display myocardial velocity profiles in each segment at rest and peak exercise. SPECT was analyzed using a 16-segment model and segments were classified as normal or showing resting or stress defects. Resting defects within segments showing normal function were attributed to attenuation. Color MDV data were compared with SPECT results, and a multivariate analysis (including exercise and SPECT results) was performed to identify the determinants of
the exercise MDV response. Patients exercised maximally (peak rate-pressure product 27.6 ⴞ 6.1 ⴛ 103), and SPECT was abnormal in 33 patients. Of the 1,333 left ventricular segments evaluable by SPECT and MDV, 1,217 segments were classified as normal, 43 showed a stress defect, and 73 a rest defect. Segmental comparison of thallium findings and MDV showed that segments with a rest defect had a lower velocity at rest and stress than normal segments (p <0.001). Segments with a stress defect had a marked reduction in peak exercise velocity and less increment in velocity than normal segments. Heart rate, functional capacity, and presence of abnormally perfused segments were independent predictors of myocardial velocity at peak exercise. Thus, color MDV correlates with independent evidence of ischemia, although it is also influenced by exercise capacity and left ventricular function. This technique may permit a feasible approach to quantitation of exercise echocardiography. 䊚2000 by Excerpta Medica, Inc. (Am J Cardiol 2000;85:536 –542)
uring the past decade, exercise echocardiography has evolved as a cost-effective alternative to nuD clear techniques for detecting and managing coronary
experimental protocol.7 In patients, MDV has also been used to quantify changes during dobutamine echocardiography,8 –12 and may be used to recognize ischemic segments.11 However, MDV has only been compared with results of wall motion scoring,11 and has not been compared with other techniques used for the diagnosis of coronary disease. Moreover, the use of exercise rather than dobutamine stress permits the assessment of exercise capacity, and constitutes a more physiologic stress, but the integration of MDV with exercise echocardiography is not well defined. The aims of this study were to validate the use of MDV during exercise echocardiography by comparison with single-photon emission computed tomography (SPECT) and to analyze the determinants of peak MDV at exercise.
disease with comparable accuracy.1,2 However, subjectivity in the interpretation of wall motion abnormalities3 and the need for adequate training4 are frequently cited as limitations to the use of this modality. Several attempts have been made to quantify regional function during stress echocardiography.5,6 The recent development of tissue Doppler measurement of myocardial Doppler velocity (MDV) has introduced a parameter that is relatively independent of two-dimensional (2-D) echo image quality, which has been used to quantify changes in regional function during an From the Department of Cardiology, Cleveland Clinic Foundation, Cleveland, Ohio; and the University of Queensland, Brisbane, Australia. This study was supported in part by the Clive and Vera Ramaciotti Foundation, Sydney, Australia. Manuscript received July 21, 1999; revised manuscript received and accepted October 7, 1999. Address for reprints: Thomas H. Marwick, MD, University Department of Medicine, Princess Alexandra Hospital, DH2W, Ipswich Road, Brisbane, Queensland 4012, Australia. E-mail: tmarwick@ medicine.pa.uq.edu.au.
536
©2000 by Excerpta Medica, Inc. All rights reserved. The American Journal of Cardiology Vol. 85 March 1, 2000
METHODS
Study patients: We studied 116 patients (age 60 ⫾ 12 years, 88 men) undergoing exercise SPECT for evaluation of known or suspected coronary artery disease. Patient agreement to participate in the study and availability of personnel and equipment were the 0002-9149/00/$–see front matter PII S0002-9149(99)00807-3
only selection criteria. Exercise testing was performed in 32 patients for diagnosis of coronary disease and for prognostic assessment (including follow-up of known disease) in 84 patients. The pretest probability of disease, derived from age, gender, and symptoms was 41 ⫾ 25%.13 A history of myocardial infarction was obtained in 38 patients, and 75 had previous revascularization. Most had multiple risk factors for coronary artery disease: 69 had systemic hypertension, 77 had hypercholesterolemia (defined by total cholesterol ⬎5.5 mmol/L, low-density lipoprotein cholesterol ⬎3.5 mmol/L, or treatment with lipid-lowering therapy), and 21 were diabetic. -Blocking agents were given to 45 patients, and 27 were taking calcium antagonists. Exercise dual-isotope SPECT: Rest thallium SPECT scintigraphy was usually performed in the morning preceding the stress test to identify the presence of resting defects. Images were obtained using a standard protocol immediately after injection of 1 mCi of thallium. We used a double- or triple-headed camera with a high-resolution collimator in a circular 360° orbit. All subjects exercised on a treadmill using Bruce or modified protocols selected in accordance with their age and exercise capacity. Before exercise a medical history was recorded and an intravenous line was secured. The electrocardiogram was monitored continuously during the test for ST changes and arrhythmias, and a 12-lead electrocardiogram was recorded before starting exercise, at the end of each stage, and after exercise. Blood pressure was measured at each stage and during the recovery period. At peak exercise, 20 mCi of technetium sestamibi was injected intravenously and the patient was asked to exercise for 1 more minute. Tests were symptom-limited, with the usual end points being fatigue, dyspnea, severe angina or severe ST-segment changes, decrease or exaggerated increase in blood pressure, or serious arrhythmia.14 Peak exercise images were obtained 1 hour after the end of the exercise using the same acquisition protocol as the resting images. Transaxial images were reconstructed using a Hamming filter in the short axis and horizontal and vertical long axes in a side-by-side rest-stress display. Images were interpreted using the same 16-segment model used for evaluation of MDV, and efforts were made during rotation and selection of imaging planes to reproduce the echo image planes based on “landmarks” at the junction of the right ventricular free wall and septum, the membranous septum, apex, and base. Segments were classified as normal, rest, or stress perfusion defects based on the comparison of rest and stress perfusion scan by observers blinded to clinical and exercise echocardiography data using a previously validated quantitative color scale.15 Ischemia by SPECT was defined as a stress perfusion defect (change of tracer activity by ⬎15%) in an area with normal perfusion at rest. A perfusion defect present on rest and stress perfusion scan (⬎20% less than maximum) was considered to be scar, unless resting function was normal, in which case the perfusion defect
was ascribed to attenuation and considered to be normal. Exercise echocardiography with color myocardial Doppler: Color MDV was recorded using commer-
cially available equipment (System FiVe, Vingmed Sound AB, Horten, Norway) with a 2.5-MHz transducer using standard settings for tissue Doppler.16,17 Images were obtained in long- and short-axis parasternal and apical long-axis views, 4- and 2-chamber views at rest, and as quickly as possible after completion of the treadmill test. Color MDV was acquired in real time, superimposed on 2-D images using a sector angle of 60° and a frame rate of at least 72 frames/s; the Nyquist limit was set at 19 cm/s and increased if necessary to avoid aliasing. Acquisition of MDV data did not require additional time compared with traditional 2-D echo. A complete cardiac cycle was acquired in each view at rest and immediately after exercise, and stored in digital format for off-line analysis as well as being recorded on VHS videotape. Wall motion analysis was performed by experienced observers blinded to other data. Regions showing a new or worse wall motion abnormality after exercise were identified as ischemic. Segments with abnormal wall motion at rest, without worsening after stress, were identified as infarcted. All other segments were identified as normal. MDV analysis was performed on color MDV cine loops derived from polar data. A region of interest was positioned in the middle of apical, mid- and basal segments of each wall,18 and a myocardial velocity profile was displayed throughout the cardiac cycle in each location (Figure 1). Analysis of the systolic component of the velocity profiles permitted measurement of peak systolic velocity at rest and after exercise in 12 defined segments. Statistical analysis: Results are reported as mean ⫾ SD. Continuous and discrete variables were compared using t tests and chi-square tests, respectively. Because neither SPECT nor wall motion scoring are completely accurate as indexes of disease, we defined normal segments based on normalcy with both techniques. Segmental SPECT interpretation was compared with systolic MDV by analysis of variance single-factor analysis with Bonferroni correction. A p value ⬍0.05 was considered statistically significant. A multiple linear regression model was used to assess the relative contribution of exercise variables and ischemia or scar to postexercise MDV.
RESULTS
Exercise responses: Clinical and exercise data are summarized in Table I. The mean ejection fraction was 64 ⫾ 13%. At exercise, 98 patients (85%) reached ⬎85% of the age-predicted heart rate. Twenty-seven patients complained of chest discomfort during the test; ischemic ST-segment changes occurred in 20 patients and 18 had ischemia by 2-D echocardiography. Feasibility and reproducibility: Of the 1,392 segments available, MDV was interpretable in 96%. Of the 59 segments without interpretable MDV data, 32 (54%) were located in the anterior wall. Based on the
CORONARY ARTERY DISEASE/MYOCARDIAL DOPPLER VELOCITY AND EXERCISE SPECT
537
FIGURE 1. Myocardial velocity profile through the cardiac cycle in a normal segment at rest (left) and during exercise (right). Velocity profile displays systolic velocity (S) and early (E) and late (A) diastolic components. At stress, both diastolic waves are frequently fused (right).
FIGURE 2. Color MDV 2-D echo of 2-chamber view (end-systolic freeze frames) with corresponding myocardial velocity profile in an inferior ischemic segment at rest (left, arrow shows site of ischemia) and during exercise (right). After stress, the increase in MDV is blunted (right arrow corresponds to peak systolic velocity).
reanalysis of 70 segments (rest or exercise), correlation between 2 readings of peak systolic MDV measurements was 0.92. The mean time to process the color MDV loops and interpret the velocity profiles for the 12 segments at rest and stress was 38 ⫾ 8 minutes/patient. 538 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 85
Correlation with thallium SPECT: Patients were divided into those with normal rest and stress SPECT (n ⫽ 83) and those with abnormal SPECT at rest (n ⫽ 14), at stress (n ⫽ 8), or both (n ⫽ 11). The mean left ventricular ejection fraction in the normal group was 58 ⫾ 11% and in the abnormal group 63 ⫾ 14% (p MARCH 1, 2000
TABLE I Clinical Data and Hemodynamic Parameters in Patients With and Without Normal SPECT
Age (yrs) Systemic hypertension Diabetes mellitus Hypercholesterolemia‡ Ejection fraction (%) Pretest probability of coronary artery disease (%) Complete test (⬎85% predicted peak heart rate) Incomplete test (⬍85% predicted peak heart rate) General fatigue Shortness of breath Chest pain/ST abnormalities Medical therapy  blocker Calcium channel antagonist Rest heart rate (beats/min) Systolic blood pressure rest (mm Hg) Diastolic blood pressure rest (mm Hg) Rest heart pressure product (⫻ 1,000) Peak heart rate (beats/min) Systolic blood pressure peak (mm Hg) Diastolic blood pressure peak (mm Hg) Peak rate pressure product (⫻ 1,000)
Normal (n ⫽ 83)
Abnormal (n ⫽ 33)
58 ⫾ 11 46 12 55 67 ⫾ 12 39 ⫾ 24
63 ⫾ 14* 23 9 22 58 ⫾ 15† 47 ⫾ 26
74
24*
9
9
8 1 0
8 0 1
31 20 71 ⫾ 12 140 ⫾ 20
14 7 71 ⫾ 12 142 ⫾ 22
85 ⫾ 11
85 ⫾ 12
9.85 ⫾ 2.1 150 ⫾ 20 190 ⫾ 26 91 ⫾ 14
10.20 ⫾ 2.7 141 ⫾ 17* 175 ⫾ 25*
TABLE II Change in Systolic Myocardial Velocities Occurring During Exercise in Normal Segments
Rest cMDV (cm/s) Exercise cMDV (cm/s) ⌬ cMDV %⌬ cMDV
Basal (n ⫽ 185)
Mid (n ⫽ 183)
Apex (n ⫽ 113)
⫾ ⫾ ⫾ ⫾
⫾ ⫾ ⫾ ⫾
⫾ ⫾ ⫾ ⫾
5.6 10.6 4.9 91
1.3* 3.9 3.2* 7.8 2.8* 3.9 62 115
1.3* 2.2 3.0* 4.5 2.9* 2.4 116 148
1.7* 3.4* 3.7* 416
*p ⬍0.001. Results are expressed as mean ⫾ SEM. Basal ⫽ basal segments; cMDV ⫽ color myocardial Doppler velocity; Mid ⫽ midsegment; ⌬ ⫽ difference between myocardial velocity at rest and exercise.
TABLE III Comparison of Rest and Stress MDV According to SPECT Results (normal, stress defect, or rest defect) Normal (n ⫽ 1,217) Rest cMDV (cm/s) Exercise cMDV (cm/s) ⌬ cMDV (cm/s) %⌬ cMDV
4.4 8.2 3.8 99
⫾ ⫾ ⫾ ⫾
1.6 3.4 2.9 106
Stress Defect (n ⫽ 43) 4.4 6.5 2.3 76
⫾ ⫾ ⫾ ⫾
Rest Defect (n ⫽ 73)
1.7 3.5 ⫾ 1.7*‡ 3.1† 5.1 ⫾ 2.8* 2.9† 1.6 ⫾ 2.1* 109 61 ⫾ 92§
*p ⬍0.0001 versus normal; †p ⬍0.005 versus normal; ‡p ⬍0.05 versus stress defect; §p ⬍0.01 versus normal. Results are expressed as mean ⫾ SEM. Abbreviations as in Table II.
89 ⫾ 14
28.48 ⫾ 5.89 24.77 ⫾ 5.71*
*p ⬍0.05; †p ⬍0.001. ‡ Defined by total cholesterol ⬎5.5 mmol/L, low-density lipoprotein (LDL) cholesterol ⬎3.5 mmol/L, or lipid-lowering therapy.
⬍0.005). Among the 83 patients with normal rest and stress SPECT, 67 had also normal exercise echocardiography based on wall motion scoring (concordance 81%). Thirty patients were abnormal on exercise echo and SPECT and 3 patients showed only abnormal SPECT (concordance 91%). SPECT analysis in segments with color MDV measurements identified 1,217 as normal, 73 showed a resting defect compatible with scar, and 43 showed an exercise-induced perfusion defect compatible with ischemia. Among the 1,217 segments with normal rest and stress SPECT scan, 1,136 had normal wall motion on exercise echocardiography (concordance 93%). Among 116 segments showing abnormal perfusion pattern on stress SPECT, 91 had wall motion abnormalities at peak stress (concordance 78%). NORMAL SEGMENTS: In 1,217 normal SPECT segments, myocardial velocity increased by 99 ⫾ 106% at peak exercise. Based on analysis of 31 patients with normal exercise echocardiography and SPECT, without a history of coronary artery disease, resting velocity in basal segments was significantly greater than in mid- or apical segments (5.6 ⫾ 1.3 vs 3.9 ⫾ 1.3 cm/s, and 2.2 ⫾ 1.7 cm/s; p ⬍0.001). This gradient remained with exercise (10.6 ⫾ 3.2 vs 7.8 ⫾ 3.0 cm/s, and 4.5 ⫾ 3.4 cm/s; p ⬍0.001) (Table II). The velocity
increment with stress was similar in basal and midsegments (p ⫽ NS). ABNORMAL SEGMENTS: Segments with rest defects had a lower MDV than normal segments (p ⬍0.0001) at rest and stress (Table III, Figure 2). Ischemic segments (stress-induced defect) were not different from normal segments at rest but had a lower MDV at peak exercise (6.5 ⫾ 3.1 vs 8.2 ⫾ 3.4 cm/s, p ⬍0.05) (Figure 3). Segments with scar had a significantly lower MDV at rest than ischemic segments (3.5 ⫾ 1.7 vs 4.4 ⫾ 1.7 cm/s; p ⬍0.05), but peak MDV was not different (p ⫽ 0.07). The normal velocity increment with exercise was 3.8 ⫾ 2.9 cm/s in normal segments. This increment was significantly greater than the velocity increment for ischemic segments (2.3 ⫾ 2.9 cm/s, p ⬍0.005) and scar segments (1.6 ⫾ 0.3 cm/s, p ⬍0.0001). The increment in myocardial velocity was comparable for scar and ischemic segments (2.3 ⫾ 2.9 vs 1.6 ⫾ 2.1 cm/s, p ⫽ NS). The increment of MDV expressed as a proportion of resting MDV showed a wide overlap in this derived value but only rest defects differed significantly from normal segments (61 ⫾ 92% vs 99 ⫾ 106%, p ⬍0.05). Correlates of the MDV response to exercise. In univariate analysis, peak exercise MDV correlated with clinical (age, risk factors, cardiac history), exercise testing (workload, hemodynamic responses), echocardiographic (segment location and overall left ventricular function), and perfusion defect extent (p ⬍0.01). However, in a multiple linear regression model, only perfusion defect extent, exercise capacity, peak heart rate, and segment location were independent predictors of peak MDV.
CORONARY ARTERY DISEASE/MYOCARDIAL DOPPLER VELOCITY AND EXERCISE SPECT
539
FIGURE 3. Color MDV 2-D echo of 2-chamber view (end-systolic freeze frames) with corresponding myocardial velocity profile in an inferior scar segment at rest (left, arrow shows site of scar) and during exercise (right). Peak systolic velocity (right arrow) is subnormal and shows a small increment with exercise.
DISCUSSION These results show that measurement of MDV is feasible during exercise echocardiography and correlates with results of exercise SPECT. Resting MDV in segments with resting perfusion defects is significantly lower than normal. The increment of MDV induced with exercise is significantly blunted in ischemic or scar segments. Maximal heart rate and functional capacity, as well as the presence of coronary artery disease, are the major determinants of peak exercise myocardial velocities. Quantitative approaches to stress echocardiography:
Several techniques, including centerline methods,19 color kinesis,5 and automatic boundary detection6 have been proposed to help interpretation of stress echocardiography. Nevertheless, none have proved to be “user friendly,” and despite limitations of wall motion scoring, the qualitative approach remains most widely used. MDV has the benefit of being quantitative and relatively independent of 2-D image quality. MDV can be displayed in spectral patterns20,21 or color coded.16,17,22 Because heart rate decreases rapidly after exercise, we chose to use color MDV to permit rapid acquisition of all data in the postexercise period. The software used for this study allowed us to circumvent previous limitations in the use of color Doppler: the temporal resolution is improved by using a high frame-rate (⬎70 frames/s), and off-line postprocessing is easy, although still time-intensive. MDV response to exercise: Conventional interpretation of stress echocardiography focuses on endocardial excursion or wall thickening in a radial direction. The apical MDV method examines base-to-apex motion, engendered by the piston-like action of the base toward the apex, which stays quite immobile during 540 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 85
the cardiac cycle.23,24 This interrogation of longitudinal rather than radial contraction is parallel with the ultrasound beam, implying that the angle dependence of Doppler is not a major limitation to accuracy. Indeed, previous stress MDV studies,9 –11,25,26 including our own,12,27 using the apical window have shown that the velocity in the basal and midzones corresponds closely to the qualitative assessment of regional function in the radial direction. Nonetheless, this correlation may merely reflect radial thickening due to longitudinal shortening, so that correlation with an independent marker of ischemia is an important development. In experimental settings, MDV has been shown to accurately reflect changes in regional and global left ventricular function induced with dobutamine, esmolol, or ischemia.7,25 In humans, induced ischemia during percutaneous transluminal coronary angioplasty balloon inflation is associated with a marked decrease in MDV and followed by recovery after balloon deflation.8 Likewise, the velocity of normal segments increases in response to dobutamine infusion,26 or exercise,27 and ischemic segments have a lower velocity response.11,12 From the apical views, the subnormal velocity increment of ischemic segments to exercise may reflect the susceptibility of longitudinal subendocardial fibers to ischemia. Comparison of echo and scintigraphy: Nuclear techniques are frequently considered as an optimal test for diagnosis of coronary artery disease because of wide experience, and because interpretation is based on objective criteria and quantification. Compared with scintigraphy, stress echo is a more recently introduced technique and current interpretation is based on subjective criteria. Although studies comparing both techMARCH 1, 2000
niques have shown similar overall accuracy,1,2 some discrepancies between the tests are inevitable, because they assess different aspects of the physiology of coronary artery disease: stress echo considers regional function and SPECT evaluates relative perfusion. For example, some segments that appear akinetic or severely hypokinetic due to necrosis affecting the subendocardial layers have a low velocity at rest, but have normal perfusion on SPECT due to the surrounding area of normal tissue. Moreover, SPECT is not a “gold standard” for the diagnosis of coronary artery disease. Segments with resting perfusion defects may show normal metabolism using positron emission tomography.28 Artifacts can play an important role in classification of segments and may lead to overestimation of the presence of scar. For this reason, perfusion data were integrated with resting wall motion, and apparent “fixed defects” in the setting of normal function were considered to be due to attenuation, and classified as normal. Study limitations: The apex only demonstrates slight motion during the cardiac cycle, so apical velocities are very low and show high variability. Moreover, the complex myocardial architecture of the heart influences the MDV that can be developed in different segments, and a large population (including a normal group) is needed to designate normal and abnormal ranges in each segment. The regional variability of MDV and problems with assessing the apex are important limitations of this technique, and we would therefore propose MDV to be used more as an adjunct to help interpretation of exercise echocardiography rather than a substitute for subjective interpretation. MDV measures velocity relative to the transducer, including velocity due to myocardial thickening, translation of the heart, and motion due to breathing. The latter can be minimized by taking images during apnea. The relative contribution of translation is unknown, although the presence of measurable velocities during ischemia or in the presence of scar suggests that this is a significant influence on MDV. Thus, the velocity increment with stress is colored by tethering of adjacent segments. The effects of translation and tethering may be reduced or removed by the use of myocardial velocity gradient rather than MDV.29 The gradient technique has been used in anterior and posterior segments in the parasternal view; we did not use this approach because of our plan to compare longaxis function in the apical view. Diastolic function is known to be altered at an early stage by ischemia. Unfortunately, the design of this exercise study only permitted data acquisition after exercise, at which time tachycardia inhibits the ability to separate E and A waves. In our previous experience,12 the feasibility of analyzing diastolic parameters was ⬍30%. However, analysis of MDV at submaximal heart rates may make regional diastolic assessment more feasible. This study is based on matching the MDV and SPECT results in the same myocardial segments, and the design is dependent on being sure that adequate matching between SPECT and echo segments has
occurred. We have developed considerable experience with this process, using landmarks to match the location of individual segments. However, it is expected that some discrepancies between the techniques arise from a mismatching of the segments. 1. Quinones MA, Verani MS, Haichin RM, Mahmarian JJ, Suarez J, Zoghbi WA.
Exercise echocardiography versus 201Tl single-photon emission computed tomography in evaluation of coronary artery disease. Analysis of 292 patients. Circulation 1992;85:1026 –1031. 2. Fleischmann KE, Hunink MG, Kuntz KM, Douglas PS. Exercise echocardiography or exercise SPECT imaging? A meta-analysis of diagnostic test performance, JAMA 1998;280:913–920. 3. Hoffman R, Lethen H, Marwick T, Arnese M, Fioretti P, Pingitore A, Picano E, Buck T, Erbel R, Flachskampf FA, Hanrath P. Analysis of interinstitutional observer agrement in interpretation of dobutamine stress echocardiograms. J Am Coll Cardiol 1996;27:330 –336. 4. Picano E, Lattanzi F, Orlandini A, Marini C, L’Abbate A. Stress echocardiography and the human factor: the importance being expert. J Am Coll Cardiol 1991;17:666 – 669. 5. Lang RM, Vignon P, Weinert L,. Echocardiographic quantification of regional left ventricular wall motion with color kinesis. Circulation 1996;93:1877–1885. 6. Perez JE Waggoner AD, Davila Roman VG, Cardona H, Miller JG. On line quantification of ventricular function during dobutamine stress echocardiography. Eur Heart J 1992;13:1669 –1676. 7. Gorcsan J III, Strum DP, Mandarino WA, Gulati VK, Pinsky MR. Quantitative assessment of alterations in regional left ventricular contractility with color-coded tissue Doppler echocardiography. Comparison with sonomicrometry and pressure-volume relations. Circulation 1997;95:2423–2433. 8. Bach DS, Armstrong WF, Donovan CL, Muller DWM. Quantitative Doppler tissue imaging for assessment of myocardial velocities during transient ischemia and reperfusion. Am Heart J 1996;132:721–725. 9. Fontanet HL Pueblo JA, Davis MG, Lockeby M, Thannikary L, Schoken DD. Quantitative dobutamine stress echocardiography utilizing Doppler tissue imaging. J Am Coll Cardiol 1996;(suppl):64A: 10. Von Bibra H, Tuchnintz A, Firscke C, Schu¨hlen H, Scho¨ming A. Doppler tissue imaging of the left ventricular myocardium-initial results during pharmacologic stress. J Am Coll Cardiol 1995;(suppl):57A: 11. Katz WE, Gulati VK, Mahler CM, Gorcsan J III. Quantitative evaluation of the segmental left ventricular response to dobutamine stress by tissue Doppler echocardiography. Am J Cardiol 1997;79:1036 –1042. 12. Yamada E, Garcia M, Thomas JD, Marwick TH. Myocardial Doppler velocity imaging—a quantitative technique for interpretation of dobutamine echocardiography. Am J Cardiol 1998;82:806 – 809. 13. Diamond GA, Forester J. Analysis of probability as an aid in the clinical diagnosis of coronary artery disease. N Engl J Med 1979;300:1350 –1358. 14. Ellestad M, ed. Stress testing protocol. In: Stress Testing. Philadelphia: FA Davis, 1986:157–186. 15. Go RT, Marwick TH, MacIntyre WJ, Saha GB, Neumann DR, Underwood DA, Simpfendorfer CC. A prospective comparison of rubidium-82 PET and thallium-201 SPECT myocardial perfusion imaging utilizing a single dipyridamole stress in the diagnosis of coronary artery disease. J Nucl Med 1990;12:1899 – 1905 16. Sutherland GR, Stewart MJ, Groundstroem KW, Moran CM, Fleming A, Guell-Peris FJ, Riemersma RA, Fenn LN, Fox KA, McDicken WN. Color Doppler myocardial imaging: a new technique for the assessment of myocardial function. J Am Soc Echo 1994;7:441– 458. 17. Wilkenshoff UM, Sovany A, Wigstrom L, Olstad B, Lindstrom L, Engvall J, Janerot-Sjoberg B, Wranne B, Hatle L, Sutherland GR. Regional mean systolic myocardial velocity estimation by real-time color Doppler myocardial imaging: a new technique for quantifying regional systolic function. J Am Soc Echo 1998; 11:683– 692. 18. Schiller NB, Shah PM, Crawford M, Demaria A, Devereux R, Feigenbaum H, Gutgesell H, Reichek N, Sahn D, Schnittger I. Recommendations for quantification of the left ventricle by two dimensional echocardiography. American Society of Echo Committee on Standards, Subcommittee on Quantitation of Two Dimensional Echocardiograms. J Am Soc Echo 1989;2:258 –267. 19. Sheehan FH, Bolson EL, Dodge HT, Mathey DG, Schofer J, Woo HW. Advantages and applications of the centerline method for characterizing regional ventricular function. Circulation 1986;74:293–305. 20. Garcia MJ, Rodriguez L, Ares M, Griffin BP, Klein AL, Stewart WJ, Thomas JD. Myocardial wall velocity assessment by pulsed Doppler tissue imaging: characteristic findings in normal subjects. Am Heart Journal 1996;132:648 – 656. 21. Hada Y, Itoh N, Tohyo Y, Yonekura K, Tamiya E, Kiritani H. Intramyocardial pulsed Doppler echocardiography as a new modality for evaluation of left ventricular wall motion: assessment in normal subjects. J Cardiol 1996; 28:85–92. 22. Donovan CL, Armstrong WF, Bach DS. Quantitative Doppler tissue imaging
CORONARY ARTERY DISEASE/MYOCARDIAL DOPPLER VELOCITY AND EXERCISE SPECT
541
of the left ventricular myocardium: validation in normal subjects. Am Heart J 1995;130:100 –104. 23. Jones CJ, Raposo L, Gibson DJ. Functional importance of the long axis dynamic of the human left ventricle. Br Heart J 1990;63:215–220. 24. Adam M, Hoglund C, Thorstand C. Longitudinal systolic shortening of the left ventricle an echocardiographic study in subjects with and without preserved global function. Clin Physiol 1992;12:443– 452. 25. Derumeaux G, Ovize M, Loufoua J, Andre-Fouet X, Minaire Y, Cribier A, Letac B. Doppler tissue imaging quantitates regional wall motion during myocardial ischemia and reperfusion. Circulation 1998;97:1970 –1977. 26. Gorcsan J III, Deswal A, Mankad S, Mandarino WA, Mahler CM, Yamazaki N, Katz WE. Quantification of the myocardial response to low-dose dobutamine
542 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 85
using tissue Doppler echocardiographic measures of velocity and velocity gradient. Am J Cardiol 1998;81:615– 623. 27. Pasquet A, Armstrong G, Beachler L, Lauer M, Marwick TH. Analysis of segmental myocardial Doppler velocity as a quantitative adjunct to exercise echocardiography. J Am Soc Echo 1999;12:901–912. 28. Brunken R, Schwaiger M, Grover-McKay M, Phelps ME, Tillisch J, Schelbert H. Positron emission tomography detect tissue metabolic activity in myocardial segments with persistent thallium perfusion defects. J Am Coll Cardiol 1987;10:557–567. 29. Tsutsui H, Uematsu M, Shimizu H, Yamagishi M, Tanaka N, Matsuda H, Miyatake K. Comparative usefulness of myocardial velocity gradient in detecting ischemic myocardium by a dobutamine challenge. J Am Coll Cardiol 1998;31: 89 –93.
MARCH 1, 2000