comparison of two-dimensional echocardiography and pulsed doppler tissue imaging during...

Comparison of Two-DimensionalEchocardiography and Pulsed Doppler TissueImaging during Dobutamine-Atropine StressTesting to Detect Coronary Artery DiseaseJesus Peteiro, M.D., Lorenzo Monserrat, M.D., Ramon Fabregas, M.D.,Jose Manuel Vazquez, M.D., Ramon Calvino, M.D., and Alfonso Castro-Beiras, M.D.

Service of Cardiology, Juan Canalejo Hospital, A Coruna, Spain

In order to compare the diagnostic accuracy of two-dimensional (2-D) echocardiography and pulsedDoppler tissue imaging (pDTI) during dobutamine-atropine stress testing (DAST) to detect signi�-cant coronary lesions, 41 patients underwent DAST (up to 40 mg/k/min of dobutamine with addi-tional atropine during submaximal heart rate responses) and coronary angiography. Pulsed Dopplertissue sampling of territories corresponding to the left anterior descending (LAD), left circum�ex(LCx), and right coronary arteries (RCAs) were performed in the apical four-chamber plus aorta andtwo-chamber apical views. The measurements were repeated at rest, at low dose (10 mg/k/min), andat peak stress. Pulsed DTI measurements included peak early systolic (Vs), peak early diastolic (Ve),and peak late diastolic (Va) velocities. Harmonic 2-D echocardiography was recorded at rest, lowdose, peak stress, and recovery, and compared with pDTI assessment. Positive 2-D echocardiographywas considered as infarction or ischemic response. The results were evaluated for the prediction ofsigni�cant coronary stenosis (50% luminal narrowing). Feasibility of pDTI was 100%, 95%, and 98%for the LAD, the LCx, and RCA territories, respectively. At rest, Vs in territories supplied by arterieswith coronary artery disease (CAD) (6.3 6 2.0 cm/sec) was not different from those without (6.6 6 2.1cm/sec). Vs increased less in territories supplied by arteries with than without CAD (75 6 107% vs102 6 69%, P 5 NS). Ve was lower in territories with CAD at rest (6.0 6 2.1 cm/sec vs 8.2 6 3.4cm/sec, P , 0.0001) and low dose (7.2 6 2.1 cm/sec vs 8.8 6 3.6 cm/sec, P , 0.01), but similar at peakstress (7.6 6 3.5 cm/sec vs 8.1 6 3.3 cm/sec). Ve increase was similar in territories with (36 6 74%)than without CAD (15 6 6 4%). Va was similar at rest and low dose in territories with and withoutCAD (9.2 6 2.7 cm/sec vs 9.1 6 2.3 cm/sec and 10.9 6 3.1 vs 10.3 6 3.6 cm/sec, respectively), butlower at peak stress in territories with CAD (13.3 6 4.6 cm/sec vs 15.3 6 4.5 cm/sec, P 5 0.05). TheVa increase was lower in territories with CAD (43 6 37% vs 77 6 72%, P , 0.05). In a territory-basedanalysis, a failure to achieve Vs $10.5 cm/sec at peak stress in the LAD and LCx, and $10.0 cm/secin the RCA territory, was found to be the more accurate limit to detect CAD in the correspondingarteries: sensitivity (95% con�dence intervals): 63% (55–71), P 5 NS vs 2-D echocardiography: 59%(51– 67); speci�city 76% (68 –84), P , 0.01 vs 2-D echocardiography: 95% (89 –100); and accuracy69% (63–75), P 5 NS vs 2-D echocardiography: 76% (70 –82). Thus, pDTI is feasible during DASTbut not more accurate than 2-D echocardiography for the detection of signi�cant CAD in a territory-based study. (ECHOCARDIOGRAPHY, Volume 18, May 2001)

pulsed Doppler tissue imaging, two-dimensional echocardiography, coronary artery disease

Although stress echocardiography has highaccuracy to detect coronary artery disease(CAD),1-5 one of its limitations is the semiquan-

titative approach to analyze regional wall-mo-tion abnormalities.

Doppler tissue imaging (DTI) is an emergingtechnique capable of a systolic and diastolicregional quantitative assessment. Previous an-imal6 and human studies7 during ischemiahave shown lack of progression of increasedsystolic velocity and changes in the regionaldiastolic pattern. An advantage of DTI over

Address for correspondence and reprint requests: JesusPeteiro, M.D., P/Ronda 5-4° izda, 15011 A Coruna,Spain. Fax: 34-981-178001, 34-981-178218; E-mail:[email protected]

Reprinted with permission fromECHOCARDIOGRAPHY, Volume 18, No. 4, May 2001

Copyright ©2001 by Futura Publishing Company, Inc., Armonk, NY 10504-0418

275ECHOCARDIOGRAPHY: A Jrnl. of CV Ultrasound & Allied Tech.Vol. 18, No. 4, 2001

two-dimensional (2-D) imaging is that theDoppler shifts signals from tissue are ofgreater amplitude.8 DTI measurements maytheoretically be possible even when 2-D imag-ing is suboptimal. We sought to compare thevalue of 2-D echocardiography and pulsed DTI(pDTI) during dobutamine-atropine stress test-ing (DAST) to detect signi�cant coronary le-sions in patients with proven or suspectedCAD.

Methods

Patients

We included 41 consecutive patients referredto the catheterization laboratory for proven orsuspected CAD. None of the patients had sig-ni�cant valvular heart disease or large prioracute myocardial infarction as de�ned by apeak creatine-kinase . 1500 U. Patients withabnormal wall motion at rest were not ex-cluded.

Dobutamine Stress Protocol

Dobutamine stress was performed with a 12-lead electrocardiogram (ECG) and blood pres-sure monitoring. Adrenergic antagonists werestopped 48 hours before the test. The stages ofdobutamine infusion were 10, 20, 30, and 40mg/k/min. Stage duration was 5 minutes at 10mg/k/min, 3 minutes at 20 and 30 mg/k/min,and 5 minutes at 40 mg/k/min. Intravenous at-ropine (0.25 mg every 1 min up to a maximumof 2 mg) was infused to achieve target heartrates if heart rate was , 85% of the patient’smaximum age-predicted heart rate (220-age)at a maximal dose of dobutamine. Endpoints ofthe test have been described in detail else-where.9,10

Two-dimensional echocardiography and pDTIwere performed using commercially availableequipment with harmonic capabilities (SequoiaC256, Acuson, Mountain View, CA; or HP-5500,Hewlett Packard, Andover, MA, USA).

Two-Dimensional Echocardiography

Two-dimensional echocardiography was per-formed in long- and short-axis parasternal andfour- and two-chamber apical views at baselineat each step of dobutamine and during recov-ery. Selected cardiac cycles at baseline, at 10mg/k/min dobutamine dosage, at peak stress,and during recovery were digitized on line andcompared off line using a quad-screen display.All cycles were reviewed by two independent

observers who were blind to all pDTI and clin-ical data. The left ventricle was divided into 16segments, according to the American Society ofEchocardiography.11 The development of newregional dysfunction (hypokinesia, akinesia, ordyskinesia), or worsening from hypokinesia toakinesia or dyskinesia, was considered an isch-emic response. The persistence of regionalbaseline dysfunction, or worsening from akine-sia to dyskinesia, was considered as infarctionwithout ischemia. Sustained improvement interritories exhibiting dysfunction at rest wasconsidered normal. A positive 2-D echocardiog-raphy result for multivessel disease was de-�ned as the presence of $ two abnormal terri-tories at rest or stress.

Pulsed Doppler Tissue Imaging

The DTI program was set to the pulsedmode. Filters were set to exclude high fre-quency signals, and the Nyquist limit was ad-justed to a velocity range of -20 to 20 cm/sec.Gains were minimized to allow for a clear tis-sue signal with minimal background noise.Pulsed DTI was recorded at baseline, 10 mg/k/min dobutamine dosage, and peak stress. Re-cordings for the left anterior descending (LAD)and left circum�ex (LCx) coronary artery ter-ritories were obtained from the apical four-chamber plus aorta view by placing the sampleat the midwall portions of the basal sites of theLV septum and lateral wall, respectively. Re-cordings for the right coronary artery (RCA)were obtained at the midwall portion of thebasal site of the posterior wall from the apicaltwo-chamber view. To achieve a proper pDTIrecording of the LCx artery territory, the trans-ducer was slightly shifted laterally to avoid anexcessive angle between the ultrasound beamand the longitudinal axis of the lateral wall.Recordings of 5–10 cardiac cycles were ac-quired digitally at a sweep speed of 50 mm/sec,and the peak systolic (Vs), peak early diastolic(Ve), and peak late diastolic (Va) velocities of3–5 cardiac cycles were measured and aver-aged off line by two investigators who wereblind to the 2-D echocardiography and clinicaldata. When two systolic waves were recorded,the earlier wave was measured. Care wastaken to discriminate systolic velocity from thevelocity during the isovolumic contraction pe-riod.

Intra- and interobserver variability in theinterpretation of pDTI data was analyzed for15 territories of �ve randomly selected patients

PETEIRO, ET AL.

276 ECHOCARDIOGRAPHY: A Jrnl. of CV Ultrasound & Allied Tech. Vol. 18, No. 4, 2001

and expressed as the difference between thetwo values divided by their mean.12

Coronary Angiography

Coronary angiography was performed byan angiographer unaware of the dobutaminestress testing results. CAD was considered tobe present if there was a 50% luminal diameternarrowing.

Statistical Analysis

Values are reported as mean 6 1 standarddeviation (SD). Student’s t-test for paired orunpaired data, when appropriate, was usedfor comparison. Qualitative data were com-pared using the chi-square test. A P value ,0.05 was considered statistically signi�cant.Sensitivity, speci�city, predictive value, andaccuracy are presented with 95% con�denceintervals.

Results

Study Group

The study group consisted of 32 men and 9women, with a mean age of 63 6 12 years,ranging from 33–78. Of the 41 patients, 37were in normal sinus rhythm, two in atrial�brillation, and two had left bundle branchblock. Eleven patients (27%) had a Q-waveacute myocardial infarction and six (15%) anonQ-wave infarction. CAD was angiographi-cally demonstrated in 32 patients; the othernine had nonsigni�cant stenosis or normal cor-

onary arteries. Rest global left ventricular (LV)dysfunction (as de�ned by a LV ejection frac-tion [EF] 45%) was present in �ve patients.Table I summarizes the clinical characteristicsof the patients.

Dobutamine-Atropine Stress Testing

Heart rate during DAST increased from 71 616 to 138 6 17 beats/min (P , 0.0001) andsystolic blood pressure from 129 6 19 to 137 627 mmHg (P , 0.05). There were no differencesin heart rate or blood pressure response be-tween patients with and without CAD. Thehighest dobutamine dose used was 40 mg/k/minfor 38 patients and 30 mg/k/min for three pa-tients. Atropine was administered to 33 pa-tients (81%). Mean dose of atropine adminis-tered was 0.65 6 0.53 mg. No major complica-tions occurred. Of the 41 tests, four wereprematurely interrupted because of the ap-pearance of limiting side effects consisting ofsevere hypotensive response (n 5 2), severenew wall-motion abnormalities (n 5 1), andintolerance (n 5 1).

Two-Dimensional Imaging

Interpretable echocardiographic images wereobtained for all patients. Interobserver concor-dance was 100% for assessment of baselinewall-motion abnormalities and 98% for detec-tion of myocardial ischemia. In case of dis-agreement, a consensus was reached.

On the basis of the response to DAST, 82territories were considered to be normal, 21

TABLE I

Baseline Patient Characteristics

CAD (n 5 32) No CAD (n 5 9) P Value

Age (y) 63 6 11 62 6 15 0.91Females (%) 8 (25%) 1 (11%) 0.65LV hypertrophy 15 (48%) 5 (55%) 0.65Hypercholesterolemia (%) 19 (59%) 5 (56%) 0.84Hypertension (%) 15 (48%) 5 (55%) 0.65Diabetes (%) 8 (25%) 0 (0%) 0.16Cigarette smoking (%) 13 (41%) 1 (11%) 0.13Family history of CAD (%) 8 (25%) 1 (11%) 0.65No. of risk factors 2.0 6 0.9 1.3 6 0.9 0.06Prior AMI (%) 14 (44%) 3 (33%) 0.71Multivessel CAD (%) 20 (63%) — —Single-vessel CAD (%) 12 (37%) — —Rest LVEF 56 6 9 61 6 4 0.12

AMI 5 acute myocardial infarction; CAD 5 coronary artery disease; LVEF 5 left ventricularejection fraction.

DTI DURING STRESS ECHOCARDIOGRAPHY


ischemic, 10 necrotic, and 10 both necrotic andischemic. One patient with a clinical history ofacute myocardial infarction who exhibited an-terior akinesia at rest but not at peak stresswas considered to be normal (this kind or re-sponse having been found to be associated withnonsigni�cant CAD), as was one patient withmoderate global dysfunction (rest LVEF 23%)who recovered at peak stress (peak LVEF55%). These patients had LAD stenosis of 25%and 75%, respectively. Vs increased more innormal than in abnormal territories (Table II).

In a territory-based analysis, 2-D echocardi-ography was abnormal in 39 of 66 territoriessupplied by a stenotic artery (sensitivity 59%),and normal in 54 of 57 territories supplied by anormal or nonsigni�cantly stenotic vessel(speci�city 95%). The positive and negativepredictive value and global accuracies were95%, 67%, and 76%, respectively.

Pulsed DTI

The quality of the pDTI tracings correspond-ing to LAD territory was considered excellentin all cases by the two observers. For the LCx,they were suboptimal at peak stress in �vecases and uninterpretable in two, whereas for

the RCA, they were uninterpretable in one.Atrial �brillation in two patients preventedmeasurement of the diastolic waves. For an-other 43 territories (35%), peak Ve and Va ve-locities could not be measured because of thefusion of the two waves at high heart rate.Intra- and interobserver variability were 2 63% and 4 6 3% for Vs measurements, and 3 63% and 4 6 3% for Ve and Va measurements,respectively.

At rest, the systolic velocities in the threeterritories did not differ signi�cantly. Atpeak stress, systolic velocities were highestin the LAD and lowest in the RCA territory(Table III). Changes in systolic and diastolicvelocities from baseline to low and peak dosein territories supplied by normal and stenoticarteries are shown in Figure 1. Vs was inde-pendent of CAD status at rest, but at low andpeak doses was higher in territories irrigatedby nonstenotic arteries. Ve was higher at restand low dobutamine dose in territories irri-gated by nonstenotic arteries, but similar atpeak stress. Va was similar at rest and lowdose but increased signi�cantly at peakstress in territories irrigated by nonstenoticarteries, as opposed to those irrigated by dis-eased vessels. Accordingly, the ratio Ve/Va

TABLE II

Rest, Peak Stress and Increase in Pulsed-Doppler Tissue Imaging Peak Systolic Velocities in Necrotic,Necrotic and Ischemic, Ischemic, and Normal Territories

Necrotic(n 5 10)

Necrotic 1Ischemic(n 5 10)

Ischemic(n 5 20)

Normal inPatients withCAD (n 5 54)

Normal inPatients without

CAD (n 5 26)

Vs rest (cm/sec) 6.8 6 2.5 6.4 6 2.5 5.9 6 1.3 6.2 6 2.0 7.2 6 2.1†‡Vs peak (cm/sec) 9.3 6 2.4 9.2 6 2.8 10.5 6 5.7 11.4 6 3.9 13.2 6 4.4¶&‡Increase (%) 47 6 47 53 6 47 95 6 168 94 6 71*q 95 6 73*\#

†P , 0.01 vs ischemic; ‡P , 0.01 vs normal in pte with CAD; ¶P , 0.01 vs necrotic; &P , 0.01 vs necrotic 1 ischemic; *P ,0.05 vs necrotic; \P , 0.05 vs ischemic; #P , 0.1 vs necrotic 1 ischemic; qP , 0.05 vs necrotic 1 ischemic; P , 0.1 vs necrotic.

TABLE III

Pulsed-Doppler Tissue Imaging Peak Systolic Velocity in cm/sec at Peak Stress in the Different TerritoriesAccording to the Presence or Absence of Coronary Artery Disease

LAD Territory LCx Territory RCA Territory

CAD 11.1 6 4.9 n 5 25 10.2 6 3.7 n 5 17 9.1 6 3.1 n 5 23No CAD 12.6 6 3.0 n 5 16 12.8 6 5.4* n 5 22 12.5 6 3.5* n 5 17

*P , 0.05 vs respective territory with CAD. Abbreviations: CAD 5 coronary artery disease; LAD 5 left anterior descending;LCx 5 left anterior circum�ex; RCA 5 right coronary artery.

PETEIRO, ET AL.


was higher at rest and low dose in territoriesirrigated by nonstenotic arteries, but similarat peak stress.

A failure to achieve Vs $10.5 cm/sec at peakstress in the LAD and LCx territories and $10.0cm/sec in the RCA territory were found to be themost accurate pDTI criteria for detection of sig-ni�cant CAD in the corresponding arteries. Withthese criteria, 41 of 65 coronary artery stenoseswere detected (sensitivity 63%). In 42 of 55 ter-ritories supplied by normal or insigni�cantlynarrowed arteries, pDTI showed peak stress ve-locities higher than the diagnostic threshold(speci�city 76%). The positive and negativepredictive values and global accuracy were76%, 64%, and 69%, respectively. Figure 2shows sensitivity, speci�city, and global accu-racy of pDTI and harmonic 2-D echocardiogra-phy for detecting signi�cant CAD in the differ-ent territories. Figures 3, 4, and 5 show pDTIrecordings obtained from three patients.

Patient-Based Analysis Comparison of 2-DImaging, pDTI, and Combined 2-DImaging and pDTI

In a patient-based analysis, 2-D imagingidenti�ed 26 of the 32 patients with CAD (sen-sitivity 81%) and eight of nine patients with-out signi�cant CAD (speci�city 89%). Vs was, 10.5 cm/sec in the LAD and LCx territory,and , 10.0 cm/sec in the RCA territorywhether the corresponding artery was stenoticin 23 out of 32 patients with CAD (sensitivity72%) and higher than these values in �ve of thenine patients without signi�cant CAD (speci-�city 56%)(Fig. 6).

The combined criteria of 2-D echocardiogra-phy positivity or blunted Vs at peak stress todetect at least one-vessel CAD achieved thehighest sensitivity (97%, P , 0.05 vs 2-D echo-cardiography, P , 0.05 vs pDTI) by detecting31 from 32 patients with CAD. In fact, Vs was

Figure 1. Change in peak early systolic (Vs), peak early diastolic (Ve), peak late diastolic (Va) regional velocities and ratioVe/Va from rest-to-low, and peak dobutamine-atropine doses according to the presence of CAD in the involved territory. *P .0.05 between groups; †P , 0.01 between groups; ‡P , 0.05 vs rest; §P , 0.001 vs rest; \P , 0.05 vs low dobutamine dose (Dob.10); ¶P , 0.001 vs low dobutamine dose (Dob. 10).



blunted in at least one territory correspondingto one stenotic vessel in �ve out of six patientswith CAD and negative 2-D imaging. However,these combined criteria were negative in only�ve out of nine patients without CAD (speci�c-ity 56%).

Discussion

The main �ndings of this study were as fol-lows:

(1) Pulsed DTI is a technically feasiblemeans of quantifying the response toDAST; and

(2) This approach did not result better thana semiquantitative visual assessment todetect CAD in the different territories.

Previous Studies

Previous studies have evaluated the regionalmyocardial velocity response to inotropic stim-ulation by either color13-16 or pulsed Dopp-ler.17-18

Color DTI allows the visualization of all ofthe myocardium in a given view, but is basi-cally a qualitative method; although quantita-tive data may be obtained off line, the velocityresolution is limited. In this study and othersusing dobutamine, rest and peak stress valueshave been lower than those measured by pDTIbecause color DTI records mean velocities, notthe peak velocities recorded by pDTI. Thus,Katz et al.16 determined a diagnostic thresholdof 5.5 cm/sec at peak stress for color DTI dis-crimination between segments that were nor-

Figure 2. Sensitivity, speci�city, and global accuracy of pDTI and 2-D echocardiography for detecting CAD in the differentterritories. Ninety-�ve percent con�dence intervals are given between parenthesis. (***, P , 0.001)

PETEIRO, ET AL.


mal and abnormal according to visual assess-ment of routine 2-D echocardiographic images.By contrast, Rambaldi et al.17 reported that inpatients without signi�cant RCA stenosis, thepDTI measured velocity of the right ventricu-lar free wall increased during DAST from 12 62 cm/sec at rest to 21 6 8 cm/sec at peak stress,whereas in patients with stenosis, the increasewas from 11 6 5 to 15 6 7 cm/sec. A 25%increase from low to peak dose had a sensitiv-ity of 82% and a speci�city of 78%. Yamada etal.18 found that in normal myocardial seg-ments, the pDTI measured velocity increasedfrom 8 6 2 to 18 6 4 cm/sec in response todobutamine, whereas in ischemic segments,the increase was from 6 6 2 to 10 6 3 cm/sec.

Present Study

Although several studies have documentedchanges in myocardial velocities with stress,we know of only one that has assessed diagnos-tic performance for detection of coronary artery

lesions, and that study was restricted to theRCA.16 In the present study, the DTI approachachieved similar sensitivity to the visual grayscale assessment, while speci�city was lower.Several explanations may account for this lackof speci�city:

(1) Individual variations in the percentageof increased velocities in response to do-butamine;

(2) Suboptimal tracings, particularly in theLCx territory, where pDTI achieved thelower speci�city; and

(3) The possibility of physiologically but notangiographically signi�cant stenosis.

However, the addition of pDTI to the 2-Dimaging assessment was very sensitive andpredictive of CAD in the individual patient. Ifthese two approaches were positive, the prob-ability of having CAD was 89%; if they werenegative, the probability was 17%.

Figure 3. pDTI recordings of the three coronary territories at rest (R), low dobutamine dose (10), and peak stress (P) in apatient with angiography showing normal coronary arteries. Peak Vs is shown in the top left of each recording. Noteprogressive increase of peak Vs in the three coronary territories from rest-to-low dobutamine dose and peak stress. Two-dimensional echocardiography was also normal. Ve 5 peak early diastolic velocity; Va 5 peak late diastolic velocity.



Systolic Myocardial Function.

The blunted myocardial velocity in territo-ries irrigated by a narrowed artery likely rep-resents a limitation of functional reserve as aresult of diminished vasodilatory capacity ofthe remote vessel.

Diastolic Myocardial Function

Transient ischemia as produced experimen-tally in animals6,19 or in patients undergoingangioplasty7 can produce signi�cant reversiblereduction of Ve and the Ve/Va ratio. Stress-producing ischemia has the same effect, al-though detection of this effect during dobut-amine stress testing is prevented in 40%17 to71%18 of cases by fusion of the early and latewaves at high heart rates. Similarly, the rela-tive magnitudes of Ve and Va are permanentlyinverted by CAD. Just as in our study, Garcia-Fernandez et al. have demonstrated a baselinemyocardial diastolic pattern of Ve , Va in myo-

cardial segments irrigated by stenotic arteriesin contrast to a normal pattern of Ve . Va insegments irrigated by normal arteries.20 How-ever, in the clinical setting, many patients re-ferred for stress testing diagnosis are elderlyand have hypertension and/or LV hypertrophy,and therefore, present baseline myocardial di-astolic abnormalities.

Since ischemia induces diastolic dysfunctionearlier than systolic, more studies analyzingmyocardial diastolic function at submaximaldoses of dobutamine are warranted.

Limitations

Forty-one percent of our patients had suf-fered acute myocardial infarction, and approx-imately the same proportion had a restingregional wall-motion abnormality in the in-volved territory, allowing diagnosis of at leastone-vessel CAD. This probably resulted inoverestimation of the sensitivity of 2-D echo-

Figure 4. pDTI recordings of the three coronary territories at rest (R), low dobutamine dose (10), and peak stress (P) in apatient with angiography showing three vessel coronary artery proximal stenosis (LAD 60%, LCx 75%, and RCA 100%). pDTI�ndings correlated with angiography. Peak Vs is shown in the top left of each recording. There was peak early Vs increase fromrest-to-low dobutamine dose but decrease at peak stress. Two-dimensional echocardiography showed ischemia in the LCx andRCA, whereas it was normal in the LAD territory.

PETEIRO, ET AL.


cardiography, although the same holds for themajority of studies evaluating stress echocar-diography.

A limitation of the DTI technique is that theangle of incidence of the ultrasound beam af-fects Doppler velocity calculations.21 The veloc-ity of myocardial motion that is not parallel tothe ultrasound beam is underestimated. TheLCx territory is particularly susceptible to thissource of inaccuracies.

DTI measurements along the long-axis view,the view used in this study, re�ect shorteningof longitudinal rather than circumferential �-bers.22 LV wall thickening during ventricularsystole results from the shortening of both lon-gitudinal and circumferential �bers. However,

Figure 5. pDTI recordings of the three coronary territories at rest (R), low dobutamine dose (10), and peak stress (P) in apatient with angiography showing proximal and severe (100% luminal narrowing) stenosis on the LCx and RCA. pDTI�ndings correlated with angiography. Peak Vs is shown in the top left of each recording. Note progressive increase of peak Vsin the LAD territory from rest-to-low dobutamine dose and peak stress, whereas in the LCx and RCA territories there wasincrease at low dobutamine dose but diminished peak Vs at peak stress. Two-dimensional echocardiography showed ischemiaassociated to necrosis in the LCx and RCA territories.

Figure 6. Comparison of pDTI, 2-D echocardiography,and combined 2-D echocardiography and pDTI to detectsigni�cant CAD in a patient-based analysis. *P , 0.05 vspDTI; †P , 0.05 vs 2-D echocardiography.



changes in longitudinal �bers seem to be moresensitive for evaluating LV.23

We did not evaluate DTI assessment for pa-tients with proximal vs distal stenoses or formoderate vs severe stenoses since the majorityof our patients had severe proximal coronaryartery lesions. It is possible that the diagnosticaccuracy of DTI might be lower for distal ste-nosis or for less severe stenosis.

Limitations regarding coronary angiographyinclude the lack of a quantitative approach.However, quantitative coronary angiography isnot routinely used for clinical decision making.

Clinical Implications

According to the results of this study, pDTIshould not be routinely performed during do-butamine stress echocardiography. However, itmay be useful in addition to 2-D imaging whenmaximal sensitivity is looked for to excludeCAD.

References

1. Segar DS, Brown SE, Sawada SG, et al: Dobutaminestress echocardiography: Correlation with coronarylesion severity as determined by quantitative angiog-raphy. J Am Coll Cardiol 1992;19:1197-1202.

2. Marwick TH, Naemec JJ, Pashkow JF, et al: Accuracyand limitations of exercise echocardiography in a rou-tine clinical setting. J Am Coll Cardiol 1992;19:74-81.

3. Hecht HS, De Bord L, Shaw R, et al: Digital supinebicycle stress echocardiography: A new technique forevaluating coronary artery disease. J Am Coll Cardiol1993;21:950-956.

4. Ryan T, Segar DS, Sawada SG, et al: Detection ofcoronary artery disease with upright bicycle exerciseechocardiography. J Am Soc Echocardiogr 1993;6:186-189.

5. Dionisopoulos PN, Collins JD, Smart SC, et al: Thevalue of dobutamine stress echocardiography for thedetection of coronary artery disease in women. J AmSoc Echocardiogr 1997;10:811-817.

6. Derumeaux G, Ovize M, Loufoua J, et al: Dopplertissue imaging quantitates regional wall motion dur-ing myocardial ischemia and reperfusion. Circulation1998;97:1970-1977.

7. Bach DS, Armstrong WF, Donovan C, et al: Quanti-tative Doppler tissue imaging for assessment of re-gional myocardial velocities during transient isch-emia and reperfusion. Am Heart J 1996;132:721-725.

8. Sutherland GR, Stewart MJ, Groundstroem KWE, etal: Color Doppler myocardial imaging: A new tech-nique for assessment of myocardial function. J AmSoc Echocardiogr 1994;7:441-458.

9. Previtali M, Poli A, Lanzarini L, et al: Dobutaminestress echocardiography for the assessment of myo-cardial viability and ischemia in acute myocardial

infarction treated with thrombolysis. Am J Cardiol1993;72:124G–130G.

10. Sicari R, Picano E, Landi P, et al: Prognostic value ofdobutamine-atropine stress echocardiography earlyafter acute myocardial infarction. J Am Coll Cardiol1997;29:254-260.

11. Bourdillon PD, Broderik TM, Sawada SG, et al: Re-gional wall motion index for infarct and noninfarctregions after reperfusion in acute myocardial infarc-tion: Comparison with global wall motion index. J AmSoc Echocardiogr 1989;9:398-407.

12. Wallerson DC, Devereaux RB: Reproducibility of quan-titative echocardiography: Factors affecting variabil-ity of imaging and Doppler measurements. Echocar-diography 1986;3:219-235.

13. Wilkenshoff UM, Sovany A, Wigstrom L, et al: Re-gional mean systolic myocardial velocity estimationby real-time color Doppler myocardial imaging: A newtechnique for quantifying regional systolic function.J Am Soc Echocardiogr 1998;11:683-692.

14. Tsutsui H, Uematsu M, Shimizu H, et al: Compara-tive usefulness of myocardial velocity gradient in de-tecting ischemic myocardium by a dobutamine chal-lenge. J Am Coll Cardiol 1998;31:89-93.

15. Gorcsan III J, Deswall A, Mankad S, et al: Quanti�-cation of the myocardial response to low-dose dobut-amine using tissue Doppler echocardiographic mea-sures of velocity and velocity gradient. Am J Cardiol1998;81:615-623.

16. Katz WE, Gulati V, Mahler Ch M, et al: Quantitativeevaluation of the segmental left ventricular responseto dobutamine stress by tissue Doppler echocardiog-raphy. Am J Cardiol 1997;79:1036-1042.

17. Rambaldi R, Poldermans D, Fioretti P, et al: Useful-ness of pulse-wave Doppler tissue sampling and do-butamine stress echocardiography for the diagnosis ofright coronary artery narrowing. Am J Cardiol 1998;81:1411-1415.

18. Yamada E, Garcia M, Thomas J, et al: MyocardialDoppler velocity imaging: A quantitative techniquefor interpretation of dobutamine echocardiography.Am J Cardiol 1998;82:806-809.

19. Garcia-Fernandez MA, Azebedo J, Puerta P, et al:Regional left ventricular diastolic dysfunction evalu-ated by tissue Doppler imaging as an earlier signal ofmyocardial ischemia: Experimental study of inducedischemia and reperfusion in pigs. J Am Coll Cardiol1996;27:299A

20. Garcia-Fernandez MA, Azebedo J, Moreno M, et al:Regional diastolic function in ischemic heart diseaseusing pulsed-wave Doppler tissue imaging. EurHeart J 1999;20:496-505.

21. Sahn DJ: Instrumentation and physical factors re-lated to visualization of stenotic and regurgitant jetsby Doppler color �ow mapping. J Am Coll Cardiol1988;12:1354-1365.

22. Oki T, Fukuda K, Tabata T, et al: Effect of an acuteincrease in afterload on left ventricular wall motionvelocity in healthy subjects. J Am Soc Echocardiogr1999;12:476-483.

23. Martin CE, Shaver JA, Thompson ME, et al: Directcorrelation of external systolic time intervals withinternal indices of left ventricular function in man.Circulation 1971;44:419-431.

PETEIRO, ET AL.


comparison of two-dimensional echocardiography and pulsed doppler tissue imaging during...

Documents