Comparison of Low-Dose Dobutamine–Gradient-Echo Magnetic Resonance Imaging and Positron Emission Tomography With [18F]Fluorodeoxyglucose in Patients With Chronic Coronary Artery Disease
A Functional and Morphological Approach to the Detection of Residual Myocardial Viability
Background There have been conflicting reports of whether substantial myocardial thinning alone as an indirect sign of myocardial scarring is sufficient evidence to exclude the presence of viable myocardium in patients with previous myocardial infarction and persisting regional left ventricular akinesia. Demonstration of a dobutamine-induced contraction reserve in postischemic viable but akinetic myocardium may serve as a direct indicator of myocardial viability. In the present study, end-diastolic wall thickness at rest and dobutamine-induced systolic wall thickening assessed by magnetic resonance imaging (MRI) were compared with corresponding [18F]fluorodeoxyglucose uptake as assessed by positron emission tomography (FDG-PET).
Methods and Results Thirty-five patients with myocardial infarction (infarct age, >4 months) and regional akinesia or dyskinesia assessed by left ventriculography underwent rest and dobutamine MRI studies (10 μg dobutamine · min−1 · kg−1) and FDG-PET followed by segmental analyses of end-diastolic wall thickness, systolic wall thickening, and FDG uptake in corresponding short-axis tomograms. Two definitions of viability, as assessed by MRI, of a segment akinetic at baseline were used: (1) end-diastolic wall thickness of ≥5.5 mm (the mean minus 2.5 SD of a healthy control group [n=21]) and (2) evidence of dobutamine-induced systolic wall thickening ≥1 mm. Segments were graded as viable by FDG-PET if FDG uptake was ≥50% of the maximum uptake in a region with normal wall motion as assessed by left ventriculography. Preserved end-diastolic wall thickness in akinetic regions was found in 17 of 35 (48%) patients at rest, and functional recovery within the infarct region was found in 19 of 35 (54%) patients during dobutamine infusion. Viability of the infarct region was indicated by FDG-PET in 23 of 35 patients (66%), yielding a diagnostic agreement between FDG uptake and myocardial morphology in 29 of 35 (83%) and between dobutamine-induced contraction reserve and FDG-PET in 31 of 35 (89%). Of 2200 segments, 482 (22%) were akinetic at rest. Of these akinetic segments, 234 (48%) had preserved end-diastolic wall thickness, 251 (52%) had a dobutamine-induced contraction reserve, and 299 (62%) were graded as viable by FDG-PET. Correlations of FDG uptake with end-diastolic wall thickness at rest (r=.48) and with dobutamine-induced wall thickening (r=.42) were similar. Comparison of segmental MRI and FDG-PET gradings indicated that dobutamine-induced wall thickening was a better predictor of residual metabolic activity (sensitivity, 81%; specificity, 95%; positive predictive accuracy, 96%) than was end-diastolic wall thickness (sensitivity, 72%; specificity, 89%; positive predictive accuracy, 91%). However, grading a segment as viable if at least one of both MRI parameters fulfilled viability criteria improved the sensitivity (88%) of MRI for FDG-PET–assessed metabolic activity without a major decrease in specificity (87%) or positive predictive accuracy (92%).
Conclusions Viable myocardium is characterized by preserved end-diastolic wall thickness and a dobutamine-inducible contraction reserve. Both parameters should be taken into account to maximize the sensitivity of MRI in the detection of regions with signs of viability on FDG-PET images.
In patients with chronic myocardial infarction and impaired left ventricular function, the distinction between an irreversible fibrotic scar causing left ventricular dysfunction and akinetic but viable myocardium has important clinical implications.1 It has been shown in pathological studies that the completely scarred myocardium of chronic infarcts is commonly associated with substantial myocardial thinning to <6 mm or even with aneurysm formation2 3 ; however, the reliability of akinesia and severely reduced end-diastolic wall thickness as criteria for the identification of completely scarred myocardium is still unclear4 5 6 7 because some investigators using [18F]fluorodeoxyglucose positron emission tomography (FDG-PET) have found residual metabolic activity, in the form of glycolysis, in myocardial regions with reduced end-diastolic wall thickness and absent systolic wall thickening.6
Recent experimental and clinical studies have suggested that postischemic myocardial dysfunction can be transiently reversed by moderate inotropic stimulation.8 9 10 11 12 Therefore, demonstration of a pharmacologically induced contraction reserve by a high-resolution imaging technique like gradient-echo magnetic resonance imaging (MRI) in basally akinetic or dyskinetic regions could be used as additional evidence for the presence of viable myocardium and could emerge as a more accurate tool for the detection of residual viable myocardium than the assessment of end-diastolic wall thickness.
The present study examines the accuracy of quantitatively assessed morphological and functional MRI parameters (end-diastolic wall thickness at rest and dobutamine-induced systolic wall thickening, respectively) for the identification of viable myocardium in patients with previous myocardial infarction and persisting left ventricular dysfunction (≥4 months since the ischemic event) as assessed by left ventriculography. These MRI parameters were compared with the results of FDG-PET, which served as the standard of reference for the differentiation of viable myocardial tissue and scar on the basis of relative FDG uptake.13 14 15
We prospectively studied 38 consecutive patients who were referred to our hospital between June 1993 and December 1993 for coronary angiography and had a documented history of previous myocardial infarction (≥4 months since the ischemic event). Only patients with regional left ventricular akinesia or dyskinesia by left ventriculography were included in the study. Patients were excluded if they had unstable angina, congestive heart failure, atrial fibrillation, a history of sustained ventricular tachycardia, or diabetes. Three patients had to be excluded from the final evaluation: in two patients there were respiratory motion artifacts preventing an accurate quantitative evaluation of the MRI studies, and in one patient PET image quality was diminished because of latent diabetes mellitus. Thirty-five patients, all men (mean age [±SD], 59±7 years; range, 44 to 73 years), remained in the study (Table 1⇓). The ECG site of myocardial infarction was anterior in 22 patients and inferior in 13. This study was approved by the Hospital Human Rights Committee (Institutional Review Board), and informed consent was obtained from every patient.
Patients underwent coronary and left ventricular angiography, FDG-PET, and dobutamine MRI studies within 10 days without intervening cardiac events. β-Blockers were withdrawn 48 hours before the test. All patients received a long-acting nitrate (Isoket Retard 120, Schwarz Pharma GmbH) before dobutamine MRI and FDG-PET studies to allow for a constant coronary artery vasodilation.
Multiple angiographic views of each coronary artery were obtained (DCI, Philips Medical Systems). Coronary artery narrowing was measured as percent diameter stenosis with electronic calipers. Regional left ventricular wall motion was visually evaluated from the left ventriculogram (right anterior oblique and left anterior oblique projections) and graded as normokinetic, hypokinetic, akinetic, or dyskinetic by two independent and experienced observers. In the cases of two patients there was disagreement, so a third observer reviewed the ventriculograms, and the majority judgment was binding. Left ventricular ejection fraction was calculated from the 30° right anterior oblique projection as the ratio of stroke volume to end-diastolic volume.
MRI studies were obtained with a commercially available 1.5-T superconducting magnet (Philips Gyroscan S15). Multislice sagittal and transaxial ECG-gated localizing spin-echo sequences were acquired to define the cardiac axes. The entire left ventricle was covered by short-axis tomograms with an interleaved acquisition technique, with imaging of two slices during each acquisition run. Slice thickness was 8 mm, and the interslice gap was 2 mm. The ECG signal was transmitted by telemetry to a remote receiver to trigger the acquisition of the images by the R wave of the ECG. The flow-compensated gradient-echo sequence used flip angles of 30° and gradient-refocused echoes with an echo time of 12 milliseconds and a repetition time of 28 milliseconds. Temporal resolution for the individual slices of the package was therefore 56 milliseconds. The acquisition matrix was 128×256, interpolated to 256×256 for display purposes. Acquisition pixel size was 2.5×1.25 mm, and the field of view was 280 mm. Measurements were repeated four times to improve the signal-to-noise ratio.
Without the patient being removed from the magnet, dobutamine (Dobutrex) was administered intravenously during continuous ECG monitoring and blood pressure recording (Boso Oscillomat, Lilly Deutschland GmbH) every 2 minutes. With a digital perfusion pump (Secura FT, B. Braun Melsungen AG), a dose of 10 μg · kg−1 · min−1 was infused into a peripheral vein beginning 5 minutes before MRI study acquisition and during acquisition. Total imaging time for both MRI studies ranged from 60 to 80 minutes and was determined by the patient’s heart rate and the number of slices needed to cover the entire left ventricle.
Analysis of MRI
Each MRI study acquired at rest was previewed in a cinematic mode from base to apex to define end-diastolic and end-systolic phases as still frames with the smallest or largest left ventricular endocardial area during the cardiac cycle. End-diastolic and end-systolic endocardial and epicardial boundaries were marked by tracing the borders on electronically magnified images. The center of mass was determined by the image-processing software of the MRI machine, and each short-axis tomogram was divided into eight segments by radii, spaced equally by 45°, emanating from the endocardial center of mass (Fig 1⇓). Mean segmental end-diastolic and end-systolic wall thicknesses were determined by dividing the myocardial area (as measured by planimeter) by the segmental perimeter. Mean systolic wall thickening was calculated by subtracting mean end-diastolic from mean systolic wall thickness.7 The same evaluation procedure was performed for the corresponding dobutamine MRI studies. Dobutamine-induced contraction reserve was calculated as dobutamine-induced systolic wall thickening minus systolic wall thickening at rest. Data on the interobserver and intraobserver variability of segmental wall thickness measurements by MRI have previously been published.4 The interobserver correlation coefficient for MRI wall thickness measurements was r=.88, and the intraobserver correlation coefficient was r=.92.
Two different definitions of viability were used. Segments were defined as viable if on MRI they showed systolic wall thickening at rest or dobutamine-induced systolic wall thickening (≥1 mm) in an akinetic or dyskinetic segment at baseline conditions. Second, they were also defined as viable if they had a mean end-diastolic wall thickness ≥5.5 mm. This threshold value was based on the mean end-diastolic wall thickness (10.5±2 mm) of a healthy control group (n=21) minus 2.5 SD.7
PET imaging was performed to assess glycolytic metabolism with FDG by a whole-body scanner (Siemens CTI ECAT EXACT 921) that had an axial field of view of 16.2 cm and was equipped with 68Ge/68Ga retractable line sources for transmission scans, as described previously.16 To increase myocardial glucose uptake, each patient drank a solution containing 50 g glucose 1 hour before the administration of FDG. Images were corrected for attenuation by use of coefficients measured by a transmission scan of 30 minutes’ duration. Emission scans (6 scans, each 5 minutes long) were started 30 minutes after injection of 370 MBq (10 mCi) FDG. The transaxial resolution was 6 mm full width at half maximum.16
FDG-PET Image Analysis
The left ventricular apex could be easily identified by both imaging techniques, and it served as the anatomic landmark for the reconstruction of PET short-axis tomograms corresponding to the respective MRI studies (Figs 2⇓ and 3⇓). In each patient, reconstructed PET tomograms (slice thickness, 10 mm) were evaluated by creating a polar map encompassing the entire left ventricle from base to apex. Each slice of this polar map was divided into eight segments using the same segmental pattern as that described for MRI (Fig 1⇑). By use of a SUN workstation, the mean FDG uptake for each segment was normalized to a myocardial segment with maximum FDG uptake. This reference segment was required to be perfused by a coronary artery with ≤70% diameter stenosis and to have normal wall motion by left ventriculography to avoid the misinterpretation of FDG uptake that could occur if the maximum FDG uptake happened to be in a viable but ischemic segment with severely reduced perfusion at rest. PET segments were defined as viable if the mean segmental FDG uptake was ≥50% of the maximum FDG uptake in a myocardial region with normal wall motion.17 18 19
All data are expressed as mean±SD. Serial changes in heart rate, systolic blood pressure, and double product during dobutamine infusion were analyzed by Student’s t test. ANOVA with Bonferroni correction was used to assess the significance level of mean FDG uptake for different MRI categories based on morphological and functional criteria. Sensitivity, specificity, and predictive accuracy of dobutamine MRI were calculated by applying standard formulas. The null hypothesis was rejected at the 95% confidence level, considering P<.05 as significant.
Of 35 patients with chronic infarction (mean infarct age, 18±16 months; range, 5 to 72 months) with interpretable dobutamine MRI and FDG-PET studies, 7 had non–Q-wave infarction (mean peak creatine kinase, 550±250 U/L) and 28 had Q-wave myocardial infarction (mean peak creatine kinase, 1850±1100 U/L). A totally occluded infarct-related vessel was found in 19 of 25 patients with Q-wave infarction and in 4 of 7 patients with non–Q-wave infarction. Six patients had one-vessel coronary artery disease, 24 had two-vessel disease, and 5 had three-vessel disease (Table 1⇑). Left ventricular ejection fraction ranged from 23% to 69% (mean, 42±16%).
Clinical and Hemodynamic Observations During Low-Dose Dobutamine MRI
The low-dose dobutamine MRI stress protocol with 10 μg dobutamine · kg−1 · min−1 was successfully completed in all patients. One patient developed frequent ventricular premature beats, and another had mild chest pain that was quickly relieved by sublingual nitroglycerin. Other side effects included palpitations (5%) and tingling or flushing sensations (7%). During dobutamine infusion, heart rate (79±15 beats per minute at rest versus 90±17 beats per minute during dobutamine), systolic blood pressure (119±17 mm Hg at rest versus 132±19 mm Hg during dobutamine), and rate-pressure product (9500±2000 mm Hg/min at rest versus 11 700±2300 mm Hg/min during dobutamine) increased significantly (P<.01).
Patient-Based Assessment of Viability by Morphological and Functional MRI Parameters and FDG Uptake
Preserved end-diastolic wall thickness ≥5.5 mm in akinetic myocardial regions at rest was identified in 17 of 35 patients (48%). The other 18 had significantly reduced end-diastolic wall thickness (<5.5 mm) in the infarct region. A dobutamine-induced contraction reserve could be identified in these regions in 19 of 35 patients (54%). There was disagreement between functional and morphological identification of myocardial viability by MRI in the cases of 4 patients; in 3 patients there was a dobutamine-induced contraction reserve despite reduction of end-diastolic wall thickness fulfilling MRI criteria for myocardial scar, and in 1 patient there was preserved end-diastolic wall thickness but persisting akinesia during dobutamine stimulation. Table 1⇑ shows the clinical and angiographic characteristics of patients with and without preserved end-diastolic wall thickness, dobutamine-induced systolic wall thickening, or both. There were no significant differences between these patient groups with respect to age, localization of infarction, and number of diseased vessels. However, in contrast to patients with Q-wave infarction, there was only 1 patient with non–Q-wave infarction who had no dobutamine-induced contraction reserve.
FDG-PET revealed the presence of viable myocardium within the infarct region (ie, a regional FDG uptake ≥50% of the maximum FDG uptake in a myocardial region with normal wall motion) in 23 of 35 patients (66%). Seventeen of these patients (74%) had preserved end-diastolic wall thickness, and 20 (87%) had dobutamine-induced systolic wall thickening in the corresponding region (Fig 3⇑). In 6 patients, remnants of viable myocardium as graded by FDG-PET were graded as scar by rest MRI (end-diastolic wall thickness <5.5 mm), and in 4 patients no contraction reserve during dobutamine MRI was seen. There were no patients with either preserved end-diastolic wall thickness or dobutamine-induced systolic wall thickening who had segments graded as scar by FDG-PET (Table 1⇑), but 3 patients had metabolic activity and a dobutamine-induced contraction reserve despite reduced end-diastolic wall thickness that fulfilled MRI criteria for myocardial scar.
Segmental Analysis of Morphological and Functional MRI Parameters in Relation to FDG Uptake
Fig 4⇓ shows the distribution of a total of 2200 corresponding MRI and PET segments. Segments with systolic wall thickening at rest (n=1718) were graded as viable; the other 482 akinetic or dyskinetic segments at rest were further evaluated for end-diastolic wall thickness and dobutamine-induced systolic wall thickening and subsequently graded as viable or scar based on the respective MRI definitions.
End-Diastolic Wall Thickness Compared With FDG Uptake
Of 482 basally akinetic or dyskinetic segments with end-diastolic wall thickness <5.5 mm, 248 (51%) were graded as scar by MRI (Fig 4⇑). FDG-PET showed scar in 163 of these 248 segments (66%). Of the 234 segments with end-diastolic wall thickness ≥5.5 mm, 214 (91%) were also graded as viable by FDG-PET. Sensitivity and specificity of end-diastolic wall thickness for FDG-PET–assessed viability were 72% and 89%, respectively. Agreement between segmental MRI grading based on end-diastolic wall thickness and PET grading in basally akinetic segments was 78% (377 of 482 segments) (Fig 5⇓). Overall correlation between segmental end-diastolic wall thickness and FDG uptake (n=2200) was r=.48 (Fig 6⇓). Correlation between segmental end-diastolic wall thickness and FDG uptake in basally akinetic segments (n=482) was r=.59. Mean segmental FDG uptake related to MRI findings is shown in Fig 7⇓.
Dobutamine-Induced Contraction Reserve Compared With FDG Uptake
In the 482 segments that were akinetic or dyskinetic at rest, dobutamine-induced systolic wall thickening was detected in 251 (52%), of which 242 (96%) were also viable as assessed by FDG-PET. In the remaining 231 segments with persisting akinesia during dobutamine stimulation, 174 (75%) were also graded as scar by FDG-PET, but 57 (25%) were graded as viable (Fig 4⇑). Sensitivity of dobutamine-induced systolic wall thickening for detecting the persistence of metabolic activity was 81%, and specificity was 95%. Agreement between segmental MRI grading based on dobutamine-induced systolic wall thickening and FDG-PET grading in basally akinetic segments was 86% (416 of 482 segments) (Fig 4⇑). Mean segmental FDG uptake related to functional MRI grading is shown in Fig 8⇓. Overall correlation between dobutamine-induced systolic wall thickening and FDG uptake (n=2200) was r=.42 (Fig 9⇓). Correlation between segmental dobutamine-induced systolic wall thickening and FDG uptake in basally akinetic segments (n=482) was r=.53.
Further differentiation of the 231 segments without a dobutamine-induced contraction reserve with respect to systolic wall thickening at rest yielded 188 akinetic and 43 dyskinetic segments (systolic wall thinning). A dobutamine-induced contraction reserve could not be observed in dyskinetic segments. Quantitative relations between FDG uptake, end-diastolic wall thickness, and dobutamine-induced systolic wall thickening are presented in Table 2⇓.
Combined Evaluation of Morphological and Functional MRI Parameters in Relation to FDG Uptake
Grading a segment as viable if at least one of both MRI parameters fulfilled the MRI criteria of myocardial viability improved the concordance between MRI and FDG-PET gradings in basally akinetic segments to 88% (Fig 4⇑). Of the 482 akinetic segments at rest, 299 (62%) were graded as viable by FDG-PET. During dobutamine infusion, 242 of these 299 segments had dobutamine-induced systolic wall thickening, and an additional 21 segments without dobutamine-induced systolic wall thickening had normal end-diastolic wall thickness (Fig 4⇑). This resulted in a sensitivity of 88% and a positive predictive accuracy of 91% for dobutamine-MRI–obtained functional and morphological parameters in comparison to FDG-PET–defined viability, but specificity was decreased slightly to 87% with the use of both parameters.
There was segmental discordance between grading by FDG-PET and by the combined dobutamine-MRI viability criteria in 60 segments (Fig 5⇑), 12% of all basally akinetic segments graded as viable by FDG-PET. Twenty-four of these segments were viable according to MRI parameters (20 had preserved end-diastolic wall thickness ≥5.5 mm, 4 had dobutamine-induced systolic wall thickening, and 6 were viable according to both MRI parameters) but were graded as scar by FDG-PET. The other 36 segments were graded as viable by FDG-PET but fulfilled neither morphological nor functional MRI criteria of viability. In most of these segments (28 of 36), FDG uptake was <60% (mean, 58±8%), close to the threshold value.
Assessment of residual myocardial viability in the setting of chronic myocardial infarction is crucial to making clinical decisions about further treatment. Because regional myocardial asynergy may arise from scar tissue or akinetic but viable myocardium,20 21 the presence of regional contraction abnormalities alone does not reliably indicate the absence of potentially salvageable myocardium. On the basis of pathological data, the preservation of end-diastolic wall thickness as assessed by MRI in akinetic myocardial regions was proposed as an indicator of viable myocardium.7 22 23 24 However, studies have challenged the reliability of myocardial morphology as assessed by MRI at rest for the identification of viable myocardium6 because metabolic activity can be observed in regions with significantly reduced end-diastolic wall thickness and no wall thickening.
The present study was designed to extend the observations of previous studies by eliciting a contraction reserve in postischemic viable but akinetic myocardium using a low-dose dobutamine infusion. MRI is a high-resolution technique with excellent definition of endocardial and epicardial borders and is therefore well suited to the quantification of dobutamine-induced systolic wall thickening.25 Because several studies have shown that the ability of FDG-PET to demonstrate glycolytic metabolic myocardial activity makes it a good test for myocardial viability,13 15 26 it was considered the reference standard in the present study.
Pharmacological Induction of a Contractile Reserve in Chronic Myocardial Infarcts
A condition of chronic contractile dysfunction in hypoperfused but viable myocardium that normalizes upon reperfusion has been defined as “hibernation.”21 In contrast to experimental models describing the short-term effects of myocardial ischemia on left ventricular function and the susceptibility of dysfunctional myocardium to inotropic stimulation,27 28 hibernation is only a clinically defined condition. Animal models of the effects of a prolonged reduction of myocardial blood flow for more than a few hours are not available.29 However, the improvement of left ventricular wall motion in regions of hibernating myocardium during inotropic stimulation has been shown to be a predictor of improved left ventricular function after revascularization.30 31 32 In particular, patients with depressed left ventricular function who demonstrate significant improvement of ejection fraction during inotropic stimulation have improved left ventricular function and a better rate of long-term survival after coronary revascularization.30 This is in agreement with a recently published echocardiographic study reporting that a dobutamine-induced contraction reserve in chronic infarcts is a good predictor of left ventricular improvement after revascularization.33
MRI Parameters for Myocardial Viability
In the present study, two definitions of viability were used. In agreement with postmortem human studies that described significant thinning of myocardium in patients with chronic transmural myocardial infarction,2 3 one possible definition of scar tissue as assessed by MRI can be based on significantly reduced end-diastolic wall thickness and absent systolic wall thickening or even systolic wall thinning.4 5 6 7 Therefore, scar tissue was morphologically defined in this study as a basally akinetic segment with mean end-diastolic wall thickness <5.5 mm. With this definition of end-diastolic wall thickness, the segmental cut-off values are in good agreement with measurements from hearts examined at autopsy, which showed that chronic transmural scar is less than 6 mm thick.2 The second definition was based on the measurement of dobutamine-induced systolic wall thickening in basally akinetic segments. Furthermore, both definitions were combined to grade each segment as viable if at least one of both MRI viability definitions was positive.
FDG-PET is considered to be the reference method for the noninvasive identification of viable myocardium in patients with a compromised left ventricle. The preserved glycolytic activity as estimated by regional FDG uptake in myocardial regions with impaired function has been reported to be accurate for the differentiation of viable myocardial tissue from scar.13 15 18 19 In this study, segments with an FDG uptake ≥50% of the maximal FDG uptake in a myocardial region with normal wall motion were defined as viable by PET. This corresponds to the common PET criteria for viability: FDG uptake ≥50% of that observed in an area with normal blood flow as measured by [13N]ammonia or [15O]water, or normal myocardial function.15 18 19 To overcome the problem of falsely relating regional FDG uptake to a maximal uptake occurring in an ischemic segment (a PET flow tracer was not available), the reference segment was assigned to a region with normal wall motion and perfusion by a vessel with ≤70% diameter stenosis (ie, with normal perfusion at rest).
Detection of Viability by Dobutamine-Induced Systolic Wall Thickening
Dobutamine-induced functional recovery was observed in 19 of 35 patients (54%) with chronic myocardial infarction and severe wall motion abnormalities as assessed by left ventriculography. Comparison of dobutamine MRI and FDG-PET yielded agreement about the identification of residual myocardial viability in the infarct zone in 31 of 35 patients (89%). In the only other study using dobutamine MRI to assess viability, an increase in systolic wall thickening of >20% was the criterion that indicated viability in 22 of 28 patients (78%) with acute infarcts and regional akinesia as determined by left ventriculography.34 Dobutamine MRI grading and grading according to a thallium 201 reinjection technique that served as the standard of reference corresponded in 86% of patients. The percentage of patients with residual myocardial viability (78%) was higher than in our study, probably because of patient selection; most patients were studied during the early postinfarction period after thrombolysis, so patients with stunned myocardium were included in this cohort.34
Cigarroa et al33 took an approach similar to ours to the diagnosis of residual viability; dobutamine contraction reserve was assessed in patients with chronic myocardial infarction by use of transthoracic echocardiography. In 24 of 49 patients (49%) with chronic myocardial infarction, there was a dobutamine-induced contraction reserve. In 25 patients who underwent revascularization, 9 of 11 (85%) with a contraction reserve had improved systolic wall thickening after revascularization and 12 of 14 patients (86%) without a contraction reserve did not improve. Although results after revascularization are not yet available for our patients, viability was detected by dobutamine MRI in a similar proportion (54%) of patients.
Segmental comparison of dobutamine MRI and FDG-PET showed metabolic activity in 57 segments without a dobutamine contraction reserve. Thus, metabolic activity as assessed by FDG-PET did not always correspond to reversible dysfunction. This finding is in agreement with animal models in which it was demonstrated that myocardial regions may become incapable of contracting if a threshold percentage of transmural damage is exceeded, even if some viable myocardium remains.35 Therefore, FDG-PET may detect metabolic activity in regions with severely impaired left ventricular function in which only small remnants of viable myocardium have survived. These regions may not be able to produce a detectable amount of contraction during dobutamine stimulation or after revascularization.13 36 Therefore, the detection of such small remnants of viable myocardium may not be clinically relevant.
Assessment of Viability by Preserved End-Diastolic Wall Thickness
Comparison of end-diastolic wall thickness and FDG uptake in the present study yielded corresponding gradings in 78% of basally akinetic or dyskinetic segments and a mismatch in 105 (22%) segments. There was a fair overall correlation (r=.48) between end-diastolic wall thickness and FDG uptake, in contrast to the weak overall correlation of r=.17 reported by Perrone-Filardi et al.6 In that study, metabolic activity was present in many regions, fulfilling morphological MRI criteria of myocardial scar (with a positive predictive accuracy—the ability of end-diastolic wall thickness <8 mm to predict the absence of metabolic activity in akinetic or dyskinetic regions—of only 55%), and the authors concluded that MRI was not useful in the differentiation between scarred and normal or viable myocardial tissue. There are several explanations for the weaker correlation between end-diastolic wall thickness and FDG uptake in their study than in the present one. First, it is possible that MRI could not accurately measure wall thickness because a spin-echo technique with a short echo time was used, which may have caused difficulties in distinguishing between intracavitary flow signal and myocardium.37 Second, it is not clear whether patients with acute and chronic infarcts were included in the study by Perrone-Filardi et al, possibly even some with reperfused myocardial infarction in the subacute phase, which often exhibits myocardial necrosis without wall thinning or aneurysm formation. If this was the case, the results would merely confirm that MRI cannot identify scar in the acute phase of myocardial infarction on the basis of diastolic wall thinning because sufficient time does not elapse since the index event to permit wall thinning to occur. To circumvent these problems, only patients with myocardial infarctions older than 4 months were included in the present study, and wall thickness measurements were made from short-axis gradient-echo MRI studies. Interestingly, in another study that included 25 patients with chronic coronary artery disease and left ventricular dysfunction, Perrone-Filardi et al38 reported a significant difference in the end-diastolic wall thickness of segments with moderate and severely reduced FDG uptake, which is in agreement with our findings.
Assessment of Viability by Preserved End-Diastolic Wall Thickness or Dobutamine-Induced Systolic Wall Thickening
During dobutamine infusion, 242 of 299 segments graded as viable by FDG-PET had dobutamine-induced systolic wall thickening, and an additional 21 segments without dobutamine-induced systolic wall thickening had normal end-diastolic wall thickness. These 21 segments may represent regions in which FDG-PET detects metabolic activity that is sufficient only to maintain structural integrity without being able to produce a detectable amount of contraction under dobutamine stimulation.13 36 If this is the case, detection of preserved end-diastolic wall thickness is only of additional benefit for the identification of residual viability as measured by FDG-PET. However, these segments may not exhibit improved wall thickening after revascularization. On the other hand, 4 segments with a dobutamine-induced contraction reserve and 20 segments with normal end-diastolic wall thickness as assessed by MRI were graded as scar by FDG-PET; this could be due to unavoidable technical problems like rotational misalignment or incorrect matching of short-axis tomograms. The distribution of segments graded as scar by FDG-PET but viable by dobutamine MRI did not correspond to a specific myocardial region or patient subgroup.
There are some limitations of the imaging techniques used in the present study that merit comment. Standard MRI techniques do not allow real-time imaging for on-line assessment of a pharmacologically induced contraction reserve, although this may become possible in the future with improved echo-planar techniques. Moreover, quantitative assessment of wall thickness at rest and after dobutamine infusion is time-consuming and difficult to use for routine clinical purposes. However, faster imaging techniques39 used in conjunction with semiautomated analysis of wall thickening changes between rest and dobutamine studies40 may soon facilitate and promote the clinical use of MRI. Limitations of FDG-PET for routine use in the evaluation of myocardial viability include problematic evaluation of FDG uptake in diabetic patients, radiation exposure, even higher cost than MRI, and limited availability.
Limitations of the study include the small number of patients, which did not permit a gender-specific differentiation of the results or a comparison between patients with Q-wave or non–Q-wave infarction. Disagreements between dobutamine MRI and FDG-PET findings may be partly ascribed to misalignment of segments due to rotational effects during the heart cycle assessed by MRI and the nontriggered PET data acquisition.
Viable myocardium is characterized by preserved end-diastolic wall thickness and a dobutamine-inducible contraction reserve. MRI is well suited to depiction and quantification of these parameters. Both end-diastolic wall thickness and pharmacologically induced contraction reserve should be taken into account to maximize the sensitivity of MRI for the detection of regions with signs of viability on FDG-PET images. Viability demonstrated by MRI as a contraction reserve in akinetic myocardium at baseline conditions may be more predictive of recovery after revascularization than the detection of myocardial glycolytic activity by FDG-PET because in MRI the potential functional competence of the myocardium is demonstrated. Therefore, comparative assessment of the prognostic value of a dobutamine-inducible contraction reserve and the presence of glycolytic activity for the postrevascularization outcome is clearly warranted.
- Received July 26, 1994.
- Revision received September 19, 1994.
- Accepted September 28, 1994.
- Copyright © 1995 by American Heart Association
Gould KL. Myocardial viability: what does it mean and how to measure it? Circulation. 1991;83:333-335.
Baer FM, Smolarz K, Jungehülsing M, Beckwilm J, Theissen P, Sechtem U, Schicha H, Hilger HH. Chronic myocardial infarction: assessment of morphology, function and perfusion by gradient-echo magnetic resonance imaging and 99mTc-methoxyisobutyl-isonitrile-SPECT. Am Heart J. 1992;123:636-645.
Perrone-Filardi P, Bacharach SL, Dilsizian V, Maurea S, Marin NJ, Arrighi JA, Frank JA, Bonow RO. Metabolic evidence of viable myocardium in regions with reduced wall thickness and absent wall thickening in patients with chronic ischemic left ventricular dysfunction. J Am Coll Cardiol. 1992;20:161-168.
Baer FM, Smolarz K, Theissen P, Voth E, Schicha H, Sechtem U. Regional 99mTc-methoxyisobutyl-isonitrile uptake at rest in patients with myocardial infarcts: comparison with morphological and functional parameters obtained from gradient-echo magnetic resonance imaging. Eur Heart J. 1994;15:97-107.
Mercier JC, Lando V, Kammatsuse K, Ninomiya K, Meerbaum S, Fishbein MC, Swan HJC, Ganz W. Divergent effects of inotropic stimulation on the ischemic and severely depressed reperfused myocardium. Circulation. 1982;66:397-400.
Ellis SG, Wynne J, Braunwald E, Henschke CI, Sandor T, Kloner RA. Response of reperfusion salvaged, stunned myocardium to inotropic stimulation. Am Heart J. 1983;107:13-19.
Bolli R, Zhu WX, Myers ML, Hartley CJ, Roberts R. Beta-adrenergic stimulation reverses postischemic myocardial dysfunction without producing subsequent functional deterioration. Am J Cardiol. 1985;56:946-948.
Brunken R, Tillisch J, Schwaiger M, Child JS, Marshall R, Mandelkern M, Phelbs ME, Schelbert HR. Regional perfusion, glucose metabolism and wall motion in patients with chronic electrocardiographic Q wave infarctions: evidence of persistence of viable tissue in some infarct regions by positron emission tomography. Circulation. 1986;73:951-963.
Bonow OR, Dilsizian V, Cuocolo A, Bacharach SL. Identification of viable myocardium in patients with chronic coronary artery disease and left ventricular dysfunction. Circulation. 1991;83:26-37.
Gould KL, Yoshida K, Hess MJ, Haynie M, Mullani N, Smalling RW. Myocardial metabolism of fluorodeoxyglucose compared to cell membrane integrity for the potassium analogue rubidium 82 for assessing infarct size in man by PET. J Nucl Med. 1991;32:1-9.
Rahimtoola SH. The hibernating myocardium. Am Heart J. 1989;17:211-221.
Sechtem U, Voth E, Baer FM, Schneider C, Theissen P, Schicha H. Assessment of residual viability in patients with myocardial infarction using magnetic resonance techniques. Int J Card Imaging. 1993;9:31-40.
Higgins CB, Lanzer P, Stark D, Botvinick E, Schiller NB, Crooks L, Kaufmann L, Lipton J. Imaging by nuclear magnetic resonance in patients with chronic ischemic heart disease. Circulation. 1984; 69:523-531.
Tamaki N, Yonekura Y, Yamashita K, Senda M, Saji H, Hashimoto T, Fudo T, Kambara H, Kawai C, Ban T, Konishi J. Relation of left ventricular perfusion and wall motion with metabolic activity in persistent defects on thallium-201 tomography in healed myocardial infarction. Am J Cardiol. 1988;62:202-208.
Schulz R, Guth BD, Pieper K, Martin C, Heusch G. Recruitment of inotropic reserve in moderately ischemic myocardium at the expense of metabolic recovery: a model of short-term hibernation. Circ Res. 1992;70:1282-1295.
Schulz R, Miyazaki S, Miller M, Thaulow E, Heusch G, Ross J Jr, Guth BD. Consequences of regional inotropic stimulation of ischemic myocardium on regional myocardial blood flow and function in anesthetized swine. Circ Res. 1989;64:1116-1126.
Matsuzaki M, Gallagher KP, Kemper WS, White F, Ross J Jr. Sustained regional dysfunction produced by prolonged coronary stenosis: gradual recovery after reperfusion. Circulation. 1983;68:170-182.
Nesto RW, Cohn LH, Collins JJ, Wynne J, Holman L, Cohn PF. Inotropic contractile reserve: a useful predictor of increased 5-year survival and improved postoperative left ventricular function in patients with coronary artery disease and reduced ejection fraction. Am J Cardiol. 1982;50:39-44.
Horn HR, Teichholz LE, Cohn PF, Herman MV, Gorlin R. Augmentation of left ventricular contraction pattern in coronary artery disease by an inotropic catecholamine: the epinephrine ventriculogram. Circulation. 1974;49:1063-1071.
Cigarroa CG, deFilippi 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.
Nienaber CA, Rochau T, Chatterjee T, Nicolas V. Dobutamin-Magnetresonanz-Tomographie und 201-Thallium-SPECT: Nachweis von vitalem Myokard in der Postinfarktphase. Z Kardiol. 1993;suppl 82:35. Abstract.
Lieberman AN, Weiss JL, Jugdutt BI, Becker LC, Bulkley BH, Garrison JG, Hutchins GM, Kallmann CA, Weisfeldt ML. Two-dimensional echocardiography and infarct size: relationship of regional wall motion and thickening to the extent of myocardial infarction in the dog. Circulation. 1981;63:739-746.
Bodenheimer MM, Banka VS, Hermann GA, Trout RG, Pasdar H, Helfant R. Reversible asynergy: histopathologic and electrocardiographic correlations in patients with coronary artery disease. Circulation. 1979;53:792-796.
Perrone-Filardi P, Bacharach SL, Dilsizian V, Maurea S, Frank JA, Bonow RO. Regional left ventricular wall thickening: relation to regional uptake of FDG and 201-thallium in patients with chronic coronary artery disease and left ventricular dysfunction. Circulation. 1992;86:1125-1137.
van Rugge FP, van der Wall EE, Spanderberg SJ, de Roos A, Matheijssen JAA, Zwinderman AH, van Dijkman PRM, Reiber JHC, Bruschke AVG. Magnetic resonance imaging during dobutamine stress for detection and localization of coronary artery disease. Circulation. 1994;90:127-138.