Prediction of Improvement of Regional Left Ventricular Function After Surgical Revascularization
A Comparison of Low-Dose Dobutamine Echocardiography With 201Tl Single-Photon Emission Computed Tomography
Background Although both 201Tl scintigraphy and low-dose dobutamine echocardiography (LDDE) have been proposed as effective methods of assessing myocardial viability, their relative efficacies are unknown. The aim of the present study was to compare the two imaging techniques in the prediction of improvement of regional left ventricular (LV) function after surgical revascularization.
Methods and Results Thirty-eight patients with severe chronic LV dysfunction (ejection fraction ≤40%, one or more akinetic [Ak] or severely hypokinetic [SH] segments on resting echocardiogram) who underwent uncomplicated coronary artery bypass graft surgery were studied with simultaneous dobutamine stress echocardiography and poststress reinjection 201Tl single-photon emission computed tomography (SPECT) before surgery. The Ak or SH segments were considered viable by LDDE when wall thickening improved during the infusion of 10 μg · kg−1 · min−1 dobutamine. Scintigraphic definition of viability was the presence of normal 201Tl uptake, totally reversible defect, partially reversible defect, or moderately severe fixed defect. The postoperative improvement of dyssynergic segments was determined with a rest echocardiogram 3 months after surgery. Of 608 LV segments, 169 were classified as Ak and 51 as SH on resting preoperative echocardiography. Of these, 170 were successfully revascularized. Wall motion during LDDE improved in 33 severely dyssynergic segments and was more frequent in SH than in Ak segments (19 of 44 versus 14 of 126, P<.0001). Viability was detected by 201Tl SPECT criteria in 103 SH or Ak segments. Thirty-two of the 33 segments from LDDE responders were judged viable on 201Tl SPECT, whereas 201Tl viability was also detected in 71 of 137 segments from LDDE nonresponders. The sensitivity and the specificity for the prediction of postoperative improvement of segmental wall motion were 74% (95% confidence interval [CI], 67% to 81%) and 95% (95% CI, 92% to 98%) by LDDE and 89% (95% CI, 84% to 94%) and 48% (95% 40% to 56%) by 201Tl SPECT, respectively. Positive predictive value of LDDE was higher than that of 201Tl SPECT (85%, [95% CI, 80% to 90%] versus 33% [95% CI, 26% to 40%]). Thirty-six patients had angina before and only 1 had angina 3 months after revascularization. High-dose dobutamine echocardiography demonstrated significant reduction in stress-induced ischemia (new or worsening of preexisting wall motion abnormalities) after surgery (from 163 to 23 LV segments).
Conclusions In patients with severe chronic LV dysfunction, LDDE is a good predictor of the improvement of dyssynergic segments after revascularization. Because 201Tl SPECT overestimates the probability of postoperative improvement of dyssynergic segments, LDDE should be the preferred imaging technique for preoperative assessment of these patients.
Coronary artery bypass graft surgery can improve regional and global ventricular performance and the functional status of patients with chronic left ventricular (LV) dysfunction.1 The concepts of stunned and hibernating myocardium have been advocated to explain such improvement.2 3 Reliable preoperative prediction of patients in whom regional and/or global left ventricular dysfunction will improve after revascularization would present several clinical advantages, including the appropriate referral for cardiac surgery of patients who are currently considered to be unsuitable for revascularization, the referral of patients for revascularization who would at present be considered only for cardiac transplantation, and the avoidance of cardiac surgery in patients in whom revascularization would result in no functional benefit (but would carry significant risk of perioperative morbidity and mortality).
Both 201Tl single-photon emission computed tomography (SPECT)2 3 4 5 and low-dose dobutamine echocardiography (LDDE) have been proposed as effective techniques for the evaluation of myocardial viability.6 7 8 9 10 11 Although LDDE is more widely available, it is unknown whether its efficacy and reliability equal those of 201Tl SPECT, which is more established in this role.2 3 4 5
To determine the relative merits of the two imaging techniques, we compared 201Tl SPECT and LDDE in the prediction of functional recovery in 38 patients with left ventricular dysfunction who were undergoing coronary artery bypass graft surgery. Postoperative resting echocardiography at 3 months was used to determine left ventricular improvement.
Forty-three patients with stable left ventricular dysfunction who were to undergo coronary artery bypass graft surgery fulfilled the study inclusion criteria: ejection fraction of ≤40% on contrast ventriculography, history of previous (>3 months old) myocardial infarction, one or more akinetic (Ak) or severely hypokinetic (SH) segments on preoperative resting echocardiography (16-segment left ventricular model), absence of recent episodes of unstable angina, absence of significant (>50%) left main stem stenosis, and absence of (hemodynamically) significant valvular disease.
Five patients were withdrawn from the study because of perioperative myocardial infarction (3 patients), resection of all dyssynergic segments (1 patient), or inability to graft any of the Ak or SH segments (1 patient). Thirty-eight patients constituted the final study population. Mean patient age was 59 years (range, 36 to 73 years), and 26 were men. All patients were symptomatic—36 had angina pectoris and 20 had dyspnea on effort. Mean angiographic left ventricular ejection fraction was 31% (range, 18% to 40%). One-vessel disease, defined as diameter stenosis of a major coronary artery of >50%, was present in 3 patients; 16 patients had two-vessel disease; and 19 patients had three-vessel disease. Four patients had undergone previous coronary artery bypass graft surgery. Four patients were on β-blockers during the preoperative diagnostic work-up.
Dobutamine Stress Echocardiography
The dobutamine stress test was performed as follows. A two-dimensional transthoracic echocardiogram in standard views and a 12-lead ECG were recorded with the patient at rest. Dobutamine was infused through an antecubital vein at dosages of 5 and 10 μg · kg−1 · min−1, for 3 minutes at each dose. Subsequently, three other steps from 20 to 40 μg · kg−1 · min−1 (3 minutes each) were added. Finally, atropine (up to 1 mg) was injected when 85% of the maximal heart rate had not been reached.12 The echocardiogram was monitored during the test, and the last minute of each stage was recorded on videotape. The echocardiographic images were also digitized (on optical disk [Vingmed CFM 800] or on floppy disk [Esaote Biomedica SIM 7000]) and displayed side-by-side in quadscreen format to facilitate the comparison of images at rest and after dobutamine with subsequent postoperative images. A 3-lead ECG was monitored continuously, and a 12-lead ECG was recorded every minute. Blood pressure was measured by sphygmomanometer at each 3-minute stage.
To assess the functional outcome of the dyssynergic segments, we obtained resting two-dimensional echocardiograms in all patients 3 months after cardiac surgery. In addition, high-dose dobutamine/atropine stress echocardiography was obtained in 32 patients.
Analysis of Preoperative and Postoperative Echocardiograms
The interpretation of echocardiograms was performed by two experienced observers who were blinded to the clinical, angiographic, and previous echocardiographic results of the individual patients. In a subset of 11 patients (176 segments), the interobserver and intraobserver variabilities of the classification of resting wall motion and the response to LDDE were also assessed. The assessment was based on both the digitized images displayed in a quadscreen format and a review of the images recorded on the videotape. The assessment was semiquantitative, and a 16-segment model13 was used. The wall motion, including wall thickening, of every segment was scored with a 5-point scoring system, where 1 is normal wall motion and thickening, 2 is moderately hypokinetic, 3 is SH, 4 is Ak, and 5 is dyskinetic. We defined a segment as SH in the presence of minimal wall thickening with a limited inward motion of <2 mm; as Ak in the absence of systolic wall motion and thickening and, whenever possible, confirmed by M-mode tracing; and as dyskinetic in the presence of systolic outward wall motion with thinning.
Wall thickening was primarily used for the classification of wall motion, preventing the problem of postoperative paradoxical septal motion. Also, to reduce the confounding effect of tethering from adjacent segments, segmental wall thickening was analyzed frame-by-frame during the first half of systole. Myocardial viability was judged to be present in a dyssynergic (either Ak or SH) segment when wall motion improved during the infusion of low-dose dobutamine by at least one point of the scoring system. Thus, a severe hypokinesis becoming moderately hypokinetic or systolic myocardial thickening becoming apparent in a previously Ak segment was considered a marker of viability. Myocardial ischemia was judged to be present when there was worsening by ≥1 of the segmental score. Ak and dyskinetic segments were not evaluated for this purpose.
Follow-up echocardiograms were compared with the corresponding preoperative resting images. For each segment, improvement of function was defined as a decrease of one or more grades. Moreover, we used the preoperative and postoperative wall motion score indexes (WMSIs) to evaluate the effect of revascularization on global LV function. WMSI was defined as the sum of the degrees of each segment divided by the total number of segments analyzed.
201Tl SPECT Imaging
Briefly, as described previously,14 15 16 201Tl (2 mCi) was injected intravenously 1 minute before the termination of the infusion of high-dose dobutamine (up to 40 μg · kg−1 · min−1, with the addition of atropine if there were no signs of ischemia and if 85% of the maximal heart rate had not been reached). The acquisition of the poststress SPECT imaging was started within 10 minutes after the interruption of the dobutamine infusion. All images were acquired with a Siemens Gammasonics single-head Rota Camera (Orbiter) and a low-energy, all-purpose collimator. Thirty-two projections were obtained, from left posterior oblique to right anterior oblique, with an acquisition time of 45 seconds for each projection. A Gamma 11 computer system was used to process the tomographic data. Four hours after the stress imaging, a second acquisition was performed 20 minutes following the reinjection of 1 mCi of 201Tl.
As previously described,15 16 the interpretation of the images was based on six short-axis slices, three longitudinal slices, and three transverse long-axis slices (both stress and after reinjection). The analysis was performed visually with the assistance of quantitative measurement (circumferential profiles). The same 16-segment model used for interpretation of the echocardiograms was applied for the interpretation of the SPECT images. Scintigraphic images from the short-axis and the long-axis views were matched with the echocardiographic images. Each defect was classified as fixed, partially reversible, or totally reversible. A myocardial segment was considered nonviable in the presence of a severe irreversible defect. A defect was classified as severe if the 201Tl uptake of a segment was <50% of the uptake of the “normal” segments on the quantitative circumferential profile analysis and if it was consistent with a severe visually assessed defect. Scintigraphic definition of viability was based on the presence of normal 201Tl uptake, totally reversible defect, partially reversible defect, or moderately severe fixed defect.
Continuous data are expressed as mean±SD. Univariate analysis for categorical variables was performed using the χ2 test with Yates’ correction. Differences were considered significant if the null hypothesis could be rejected at the .05 probability level. Sensitivity, specificity, and positive and negative predictive values were based on their standard definitions and are reported with the corresponding 95% confidence interval (CI). The interobserver and intraobserver variabilities of regional wall motion pattern were assessed as percent agreement and κ value.
Of a total of 608 LV segments, 169 were classified as Ak and 51 as SH. Forty-three Ak segments were excluded from the postoperative evaluation because of aneurysmectomy (n=41) or because they were not grafted (n=2). Of the 51 SH segments, 7 were excluded from the postoperative evaluation because of aneurysmectomy (n=3) or because they were not graftable (n=4).
Low-Dose Dobutamine Echocardiography
Only 14 of the 126 Ak segments that could be successfully revascularized showed the presence of wall thickening during low-dose dobutamine on preoperative echocardiogram. These 14 Ak segments were detected in 8 patients.
Wall thickening improved during the infusion of dobutamine in 19 of the 44 SH segments. These 19 segments were present in 9 patients. Thus, viability was detected more frequently in SH than in Ak segments (P<.0001).
The interobserver and intraobserver concordances of resting wall motion analysis were 84% (κ, 0.79) and 87% (κ, 0.82), respectively. The interobserver and the intraobserver concordances of the response of wall motion during LDDE also were excellent: 92% (κ, 0.84) and 94% (κ, 0.86), respectively.
High-Dose Dobutamine Echocardiography
At peak stress, new or worsening of preexisting wall motion abnormalities was detected in 163 of 576 segments (in 32 of 36 patients). Angina occurred in 28 patients during the test.
201Tl SPECT imaging indicated the presence of viable myocardium in 49 of 112 Ak regions not responding and in 14 of 14 Ak regions responding to dobutamine and in 22 of 25 SH regions not responding and in 18 of 19 SH regions responding to dobutamine (Table 1⇓). From these data, it is clear that 201Tl SPECT indicates viable myocardium more frequently than LDDE (P<.001). The Figure⇓ displays the distribution of perfusion patterns by 201Tl SPECT according to the different LDDE results.
Postoperative Clinical and Echocardiographic Data
At 3 months after surgery, only 1 patient had angina and 10 patients still complained of dyspnea on effort.
Improvement of Regional Wall Motion
Resting echocardiograms at 3-months postoperative follow-up revealed an improvement of wall motion in 38 (22%) of the 170 dyssynergic segments. The improvement was found in 22 (17%) of the 126 Ak segments (change to SH in 6 segments, to moderate hypokinesis in 13 segments, and to normal in 3 segments) and in 16 (36%) of the 44 SH segments (change to moderate hypokinesis in 8 segments and to normal in 8 segments) (P=.02).
Reduction in Dobutamine-Induced Myocardial Ischemia
Of 32 patients who underwent a high-dose dobutamine stress test during follow-up, 3 had angina at peak stress. New or worsened wall motion abnormalities were detected in 10 patients and in 23 of 512 LV segments.
Prediction of Regional Improvement
Postoperative improvement occurred in 28 of 33 segments that improved during LDDE and in only 10 of 137 segments that did not improve (Table 1⇑). Of the segments judged to be viable by LDDE, postoperative improvement occurred in 79% of SH segments and in 93% of Ak segments.
201Tl SPECT detected the presence of viable myocardium in 32 of the 33 matched segments considered viable by LDDE. Despite the frequent indication of viability by 201Tl SPECT in the Ak segments unresponsive to dobutamine (44%), improvement after surgery was found in only 8% of these segments. Similarly, wall thickening improved after revascularization in only 1 of the 25 SH regions unresponsive to dobutamine, despite signs of viability by 201Tl SPECT in 22 of these 25 segments. Table 2⇓ shows the predictive accuracy with 95% CI of the two methods for the postoperative improvement of SH and Ak segments.
Global Left Ventricular Function
The WMSI revealed that there were no significant differences before and after coronary artery bypass graft surgery in either the subset in whom myocardial viability was predicted by LDDE (13 patients) (2.6±0.5 versus 2.4±0.4) or in the entire study population (2.3±0.5 versus 2.3±0.5).
A routine coronary angiogram independent of the recurrence of symptoms was undertaken in 14 patients at 3-month follow-up. Sustained patency of the grafts to the Ak or SH segments was demonstrated in all of these patients.
The functional assessment of hibernating myocardium and the presence of viability is clinically challenging3 and of paramount importance for the selection of the most appropriate individual treatment for patients with chronic severe LV dysfunction.1
We prospectively studied a group of patients with severe chronic LV dysfunction who were candidates for surgical revascularization (1) to assess the prevalence of regional improvement of Ak and SH segments after surgical revascularization and (2) to evaluate the roles of LDDE and 201Tl SPECT for predicting such improvement. In addition, the reversibility of stress-induced myocardial ischemia was assessed by clinical judgment and high-dose dobutamine stress test.
In this series of patients, improvement of regional function after revascularization was found in 22% of Ak and SH segments. This percentage is lower than that of other series7 9 10 and might be related to the selection of patients with severely impaired LV function in a tertiary referral center with an ongoing heart transplantation program.
We have demonstrated that preoperative LDDE is both a sensitive (28 of 33 segments) and a specific (127 of 137 segments) predictor of postoperative outcome of regional myocardial function. The pattern of improvement of wall thickening in severely dyssynergic regions during the infusion of low-dose dobutamine was found to be a reliable predictor of functional recovery of wall motion after successful and uncomplicated surgical revascularization, with a positive predictive value of 85% (95% CI, 80% to 90%), whereas the pattern of Ak or SH unresponsive to low-dose dobutamine is indicative of nonviable tissue and has a negative predictive value of 93% (95% CI, 89% to 97%).
201Tl SPECT (matched for echocardiographic segments) indicated the presence of viable myocardium more frequently than LDDE (103 of 170 versus 33 of 170). However, this imaging technique appears to be less suitable than LDDE to predict the postoperative improvement of regional wall motion in patients with severe LV dysfunction. In particular, the high prevalence of viability detected before surgery by 201Tl SPECT and the low prevalence of postoperative functional improvement result in a low specificity and in a low positive predictive value (Table 2⇑).
Overestimation of myocardial viability by perfusion scintigraphy may relate to several factors. First, scintigraphy may detect islands of jeopardized vital myocardial cells of inadequate size to revert LV dysfunction despite successful revascularization. Second, tethering by scar tissue may restrict the improvement in wall motion of adjacent viable segments. Third, functional recovery may not be complete by 3 months (“embalmed myocardium”).17 Finally, since subendocardial layers play a major role in wall motion, a necrosis limited to the subendocardium may result in severe dyssynergy despite the presence of viable myocardium in the subepicardial layers.18
In our study group, there was no significant postoperative improvement of global LV function; however, bypass surgery alleviated myocardial ischemia, since the number of patients with angina and the extent of stress-induced ischemia were greatly reduced. This is consistent with the high postoperative patency rate of bypass coronary grafts and confirmed the usefulness of dobutamine stress echocardiography in the assessment of stress-induced myocardial ischemia after coronary revascularization.12
Although several studies6 7 8 11 19 have addressed the role of dobutamine echocardiography for the assessment of LV functional recovery in patients with recent myocardial infarction, few data are available regarding its predictive value for postrevascularization functional improvement.9 10 In two previous studies on postrevascularization recovery, Marzullo et al9 and Cigarroa et al10 reported a higher incidence of wall thickening during low-dose dobutamine in Ak regions (47% and 39%, respectively, compared with 11% in the present study). This discrepancy may relate to different methodologies (absence of subclassification for SH segments in their studies) and patient selection (inclusion in the present study of patients with more severe and more long-standing ventricular dysfunction, where stunned myocardium is less likely to be present). Considering the value of LDDE in predicting postoperative functional outcome, our findings are in agreement with those of the two previous reports.
First, the number of viable Ak and SH segments identified in the present study was limited despite our analysis of 608 segments both before and after coronary artery bypass graft surgery. This, however, reflects our stringent inclusion criteria and our strict method of analysis (panel review with simultaneous, quadscreen format).
Second, we arbitrarily timed the outcome of dyssynergic segments 3 months after surgical revascularization. However, it cannot be excluded that functional improvement can also occur later.
Finally, we focused on the postoperative phase of regional wall motion. We are aware that the improvement in a limited area of myocardium can be clinically not relevant to global LV function. However, this was not the primary aim of the study.
Our observations in the setting of severe chronic LV dysfunction indicate that (1) wall thickening during low-dose dobutamine in Ak segments is infrequent, (2) responsiveness of Ak and SH segments to low-dose dobutamine is both a specific and a sensitive predictor of postrevascularization functional improvement, and (3) compared with LDDE, 201Tl SPECT has an equivalent sensitivity for the prediction of postoperative myocardial functional improvement but a lower specificity. Thus, LDDE should be the preferred imaging technique for predicting the functional outcome of patients with severe LV dysfunction who are under consideration for coronary artery bypass graft surgery.
A. Elhendy is supported by the Egyptian Ministry of High Education. D. Keane is recipient of a travel grant from Peel Medical Research Trust, London, UK. We are deeply grateful to Eric Boersma, MSc, for statistical assistance. Also, we appreciate the cooperation of the referring Rijnmond cardiologists.
Reprint requests to Paolo M. Fioretti, MD, University Hospital Rotterdam-Dijkzigt, Thoraxcenter Ba 300, Dr Molewaterplein 40, 3015 GD Rotterdam, The Netherlands.
- Received October 18, 1994.
- Revision received December 8, 1994.
- Accepted December 18, 1994.
- Copyright © 1995 by American Heart Association
Bonow RO, Dilsizian V. Thallium-201 and technetium-99-m-sestamibi for assessing viable myocardium. J Nucl Med. 1992;33:815-818.
Dilsizian V, Bonow RO. Current diagnostic techniques of assessing viability in patients with hibernating and stunned myocardium. Circulation. 1993;87:1-20.
Piérard LA, De Landsheere CM, Berthe C, Rigo P, Kulbertus HE. Identification of viable myocardium by echocardiography during dobutamine infusion in patients with myocardial infarction after thrombolytic therapy: comparison with positron emission tomography. J Am Coll Cardiol. 1990;15:1021-1031.
Barilla F, Gheorghiade M, Alam M, Khaja F, Goldstein S. Low-dose dobutamine in patients with acute myocardial infarction identifies viable but not contractile myocardium and predicts the magnitude of improvement in wall motion abnormalities in response to coronary revascularization. Am Heart J. 1991;122:1522-1531.
Smart SC, Sawada S, Ryan T, Segar D, Atherton L, Berkowitz K, Bourdillon PDV, Feigenbaum H. Low-dose dobutamine echocardiography detects reversible dysfunction after thrombolytic therapy of acute myocardial infarction. Circulation. 1993;88:405-415.
Cigarroa CG, de Filippi CR, Brickner ME, Alvarez LG, Wait MA, Grayburn PA. Dobutamine stress echocardiography identifies hibernating myocardium and predicts recovery of left ventricular function after coronary revascularization. Circulation. 1993;88:430-436.
Salustri A, Elhendy A, Garyfallydis P, Ciavatti M, Cornel JH, Ten Cate FJ, Boersma E, Gemelli A, Roelandt JRTC, Fioretti PM. Prediction of improvement of ventricular function after first acute myocardial infarction using low-dose dobutamine stress echocardiography. Am J Cardiol. 1994;74:853-856.
Sawada SG, Segar DS, Ryan T, Brown SE, Dohan AM, Williams R, Fineberg NS, Armstrong WA, Feigenbaum H. Echocardiographic detection of coronary artery disease during dobutamine infusion. Circulation. 1991;83:1605-1614.
Forster T, McNeill AJ, Salustri A, Reijs AEM, El-Said EM, Roelandt JRTC, Fioretti PM. Simultaneous dobutamine stress echocardiography and technetium-99m isonitrile single-photon emission tomography in patients with suspected coronary artery disease. J Am Coll Cardiol. 1993;21:1591-1596.
Sklenar J, Ismail S, Villanueva FS, Goodman NC, Glasheen WP, Kaul S. Dobutamine echocardiography for determining the extent of myocardial salvage after reperfusion: an experimental evaluation. Circulation. 1994;90:1502-1512.
Picano E, Marzullo P, Gigli G, Reisenhofer B, Parodi O, Distante A, L’Abbate A. Identification of viable myocardium by dipyridamole-induced improvement in regional left ventricular function assessed by echocardiography in myocardial infarction and comparison with thallium scintigraphy at rest. Am J Cardiol. 1992;70:703-710.