Factors Influencing Regional Myocardial Contractile Response to Inotropic Stimulation
Analysis in Humans With Stable Ischemic Heart Disease
Background We hypothesized that the response of a myocardial segment to maximal dobutamine reflects not only maximal blood flow but also tethering, metabolic, and β-blocker status.
Methods and Results Patients with stable ischemic heart disease (n=27) had positron emission tomographic measurement of blood flow at rest and with adenosine, and echocardiography at rest and with dobutamine. Positron emission tomographic measurement of [18F]fluorodeoxyglucose myocardial distribution also was made. Adenosine blood flow in segments that contracted normally at peak dobutamine was similar to that of segments that became hypokinetic (1.06±0.72 versus 1.02±0.77 mL·g−1·min−1). Segments that became akinetic failed to augment blood flow (0.68±0.30 mL·g−1·min−1). Fluorodeoxyglucose–blood flow mismatch was more common in segments with abnormal wall motion at peak dobutamine (24 of 59, 41%) versus those that contracted normally (63 of 269, 23%; χ2, 7.40; P<.01). In patients off β-blockers, segments that contracted normally at peak dobutamine increased blood flow with adenosine (0.70±0.31 to 0.86±0.46 mL·g−1·min−1; P<.05), whereas those that became abnormal did not (0.63±0.24 to 0.65±0.19 mL·g−1·min−1; P=NS). Segments of patients on β-blockers that contracted normally at peak dobutamine increased blood flow with adenosine (0.78±0.31 to 1.10±0.70 mL·g−1·min−1; P<.05), as did segments that became abnormal (0.74±0.34 to 1.06±0.82 mL·g−1·min−1; P=NS). However, segments adjacent to ones with abnormal wall motion at rest had higher frequency of abnormal response at peak dobutamine in groups on (48% versus 16%; χ2, 14.1; P<.001) and off (51% versus 21%; χ2, 10.9; P<.01) β-blockers.
Conclusions Augmented contraction at maximal dobutamine depends not only on increased myocardial blood flow but also on tethering, metabolic, and β-blocker status. Furthermore, impaired flow reserve does not preclude a normal response to maximal dobutamine, since blood flow need not increase greatly to meet demand.
The relationship between regional myocardial blood flow and regional contractile response to inotropic stimulation is likely to be complex in patients with ischemic heart disease. Studies in animals have indicated that under basal conditions, a stepwise reduction in myocardial blood flow results in a progressive decline in contractile function at rest, although the exact nature of this relationship (eg, linear versus nonlinear) remains uncertain.1 2 3 Although the extent to which flow must increase to permit enhanced contraction in response to an inotropic stimulus has been investigated in animals,3 4 5 6 7 8 few data are available in patients with ischemic heart disease. Further, results of animal studies indicate that treatment with β-blocking drugs may greatly ameliorate or even prevent a stress-induced regional wall motion abnormality at the same level of blood flow and stress as was associated with abnormal contraction before treatment.4 5 9 Similarly, in the case of “coronary steal,” comparable reduction in endocardial blood flow may not result in a comparable degree of myocardial ischemia if the drug that causes the “steal” also has intrinsic negative inotropic effects in comparison with another drug that does not.10
Accordingly, on the basis of results of animal studies cited above, this study tests the hypothesis that regional wall motion response to catecholamine stress in humans with stable ischemic heart disease may not correlate well with the ability of a segment to augment blood flow maximally. Factors that reflect metabolic status of the myocardium, major determinants of myocardial oxygen demand, and purely mechanical forces such as tethering also are likely to play a role.
Patient Recruitment and Informed Consent
After approval was obtained from both the Radiation Safety and the Human Studies committees of the Massachusetts General Hospital, patients were recruited at our institution between April 1, 1994, and March 5, 1995. Written informed consent was obtained from all patients. Exclusion criteria included any factor that precluded adequate or safe completion of either dobutamine stress echocardiography or positron emission tomography (PET). These included unstable angina, uncontrolled left ventricular failure or atrial fibrillation, severe hypertension, Wolff-Parkinson-White syndrome, myocardial infarction within 6 days, significant aortic stenosis, severe chronic obstructive pulmonary disease, inability to lie supine for sufficient time to allow data acquisition, or the necessity for urgent surgical revascularization. A total of 41 patients were approached to participate, 32 of whom consented and were enrolled. However, the final study group was limited to 27 patients after 5 who had no segments with normal wall motion at rest were eliminated. Segments with abnormal wall motion at rest were eliminated from consideration to avoid myocardial regions with scar, stunning, or hibernation.11 12 13 14
PET imaging was performed on a whole-body tomograph (Scanditronix PC4096, GE Medical Systems) in patients after an overnight fast by a previously described protocol.15 Cardiac medications were continued as prescribed by the patient's physician, although β-blockers were withheld, when possible, for 24 hours before study. Briefly, images were acquired simultaneously in 15 contiguous sections, with center-to-center separation of 6.5 mm. After positioning in the scanner, a 10-minute transmission scan was performed to correct the emission data for attenuation. Two minutes after initiation of an infusion of adenosine (140 μg·kg−1·min−1 IV over 5 minutes), dynamic data acquisition was begun, and several seconds later, ≈25 mCi [13N]ammonia IV was administered over 30 seconds. Data were collected for the first 3 minutes at 6 seconds per frame and then at 2 minutes per frame for 6 minutes. After image acquisition, radioactivity was allowed to decay for ≈30 minutes, at which time the count rate seen by the scanner was ≈7500/s. A second injection of ≈25 mCi [13N]ammonia IV was administered, with dynamic tomographic imaging begun just before injection. Images were acquired in the same fashion as described above. The radiation dose associated with a single [13N]ammonia study was 5 mR/mCi whole body and 51 mR/mCi to the bladder (target organ). The patient's ECG and arterial pressure (Dynamap, model 845, Critikon Co) were monitored continuously during the study.
After completion of the [13N]ammonia study, the patient was given an oral glucose load (Glucola, 100 g dextrose) followed by 7.5 mCi of intravenous [18F]fluorodeoxyglucose (FDG) ≈45 minutes later. One hour later, the patient was repositioned in the scanner, and a single 15-minute image was obtained. After completion of the FDG scan, a second 10-minute transmission scan was acquired for reconstruction of the FDG study. Residual radioactivity in each patient was 30- to 40-fold less than the 69Ge pin source and thus did not interfere with the second transmission scan. The radiation dose associated with the FDG study was 39 mR/mCi whole body and 440 mR/mCi to the bladder (target organ).
Attenuation-corrected [13N]ammonia and FDG images were reconstructed with a conventional filtered back-projection algorithm as 128×128 pixel images in the transverse plane normal to the long axis of the body. The projection data were filtered with a Hanning filter to yield output resolution of 7.8 mm (full width half maximal). The [13N]ammonia scans (n=3), corresponding to the last 6 minutes of data acquisition, were summed to permit placement of a region of interest over the left ventricular cavity. The region of interest was used to generate the arterial input function for the tracer kinetic model by which regional myocardial blood flow was determined.15 The arterial input function was not corrected for recirculation of labeled-ammonia metabolites.16 A computer program written at our institution was used in conjunction with the dynamic data to generate a parametric (K1) image for rest and stress conditions.17 The images obtained provided a pixel-by-pixel representation of K1 and were used for analysis of regional myocardial blood flow.
Dobutamine echocardiography, usually on the same day but always within 24 hours of PET, was performed with continuation of all medications except β-blockers, which were withheld for 24 hours before study whenever possible. After baseline blood pressure and heart rate were obtained, two-dimensional echocardiography was performed with a 2.5-MHz transducer and commercially available scanner. Images also were digitized on-line and viewed side by side in quad-screen format (ImageVue DCR, NovaMicrosonic). Echocardiography was performed in the left lateral decubitus position and included parasternal long- and short-axis views at the base (mitral), midventricle (papillary muscle), and apical levels. Apical four- and two-chamber views were also obtained. Images were recorded at baseline and then during the final minute of each dobutamine infusion at doses of 5, 10, 20, 30, and 40 μg·kg−1·min−1. Dobutamine was given up to 40 μg·kg−1·min−1 whenever possible. End points for termination of the infusion before the 40-μg·kg−1·min−1 dose included a new wall motion abnormality in at least two contiguous myocardial segments, severe anginal pain, significant hypotension (approximate decrease in systolic blood pressure >20 mm Hg), or malignant ventricular arrhythmias. The 5- and 40-μg·kg−1·min−1 doses were given for 5 minutes each, while other doses were given for 3 minutes each.18
Three short-axis rings corresponding to the proximal, middle, and distal thirds of the left ventricle were constructed for each K1 and FDG scan (Figure⇓). Next, circular regions of interest (≈8.5-mm radius) were placed over each ring at standard areas of interest: inferoseptum, midseptum, anteroseptum, anterior, anterolateral, lateral, posterolateral, and inferior zones. In the event that the myocardial ring was too small to accommodate eight evenly spaced regions of interest, a lesser number were equally placed and data from adjacent zones were averaged to interpolate values for the intervening segment.
Identical regions of interest were placed over appropriately matched proximal, middle, and distal short-axis rings of the FDG scan. FDG uptake in each ring was normalized by first determining the myocardial region with best flow in that ring. FDG uptake in the zone with best flow was set to unity, and FDG in other zones was normalized to it. Regional myocardial blood flow was normalized according to the region with highest rest blood flow. To determine whether an FDG–blood flow “mismatch” (defined as a difference >0.34) was present, relative flow was then subtracted from relative FDG uptake.15
In accordance with echocardiographic convention, the left ventricle was divided into 20 segments for analysis. Short-axis rings at the base, middle, and distal thirds of the left ventricle were used. Rings at basal and mid left ventricular levels were divided into eight sections, each of which corresponded to those of the PET scan. The ring at the distal third of the left ventricle, however, was divided into four sections (septum, anterior, lateral, and inferior). For the purposes of this study, wall motion in the septum of the distal left ventricular ring was paired with perfusion in septal and inferoseptal PET segments of the distal third of the left ventricle. Similarly, anterior wall motion was matched with anteroseptal and anterior PET sections, lateral wall motion with anterolateral and lateral PET sections, and inferior wall motion with posterolateral and inferior PET sections of the distal left ventricular ring only.
Echocardiograms were reviewed by two experienced observers who were blinded to clinical and catheterization data. Disagreements were mediated by consensus. Regional wall motion scores were determined from review of both videotape and quad-screen-formatted, digitized, single cardiac cycles at baseline and peak-dose dobutamine. The wall motion of each segment was graded according to an ordinal scale as follows: 0, normal motion; 1, mild hypokinesis; 2, severe hypokinesis; 3, akinesis; and 4, dyskinesis.
Although it was by no means the primary purpose of the study, the sensitivity of dobutamine echocardiography in this particular data set for detection both of patients and of coronary vessels with stenoses ≥70% diameter reduction was assessed as follows. A patient was identified as having ischemic heart disease if one or more myocardial segments had a stress-induced, reversible wall motion abnormality at high-dose dobutamine. Similarly, a coronary artery was identified as having a significant stenosis (as defined above) if at least one segment in its vascular territory exhibited a reversible wall motion abnormality at high-dose dobutamine. In matching segmental wall motion abnormalities with coronary vessels, the following overlap zones were considered to apply: (1) septal and inferoseptal segments were assigned to the left anterior descending or right coronary artery, (2) the posterolateral segment to the right or left circumflex coronary artery, and (3) the anterolateral segment to the left anterior descending or left circumflex coronary artery. The inferior segment was considered a “pure” right coronary zone (right dominant circulation), the anterior and anteroseptal segments pure left anterior descending, and the lateral segment pure left circumflex coronary artery. Since all patients had ischemic heart disease and since only two had single-vessel disease, it was not possible to determine specificity in this data set. The purpose of the analysis was only to determine whether sensitivity of dobutamine echocardiography in this data set approached that reported by others (References 19 and 20, and see below).
Cardiac catheterization was performed by the Judkins technique according to standard clinical methods. Coronary artery stenosis diameter was measured by combined visual and hand-caliper technique by an experienced cardiac radiologist who was unaware of results of other imaging studies. Patients were classified as having 1-, 2-, or 3-vessel coronary artery disease on the basis of the number of coronary arteries having ≥70% luminal diameter reduction.
All data are expressed as mean±SD. Group mean values of continuous variables were compared by ANOVA with appropriate post hoc multiple comparison testing (Fisher's protected least significant difference) using commercially available software (StatView V4.0, Abacus Concepts). Contingency tables and χ2 analysis also were performed with StatView V4.0. A logistic multiple regression analysis was performed with SAS software.
The mean age of the patients was 62±8 years (range, 49 to 74 years) (Table 1⇓). Twenty-three were men and 4 women. Nineteen patients had a history of myocardial infarction, of which 5 had occurred within 1 month of the study. Infarct location was both anterior and inferior in 5, anterior alone in 5, inferior alone in 5, and non–Q-wave in 4. Four patients had previous coronary bypass surgery (n=3) or coronary angioplasty (n=1). Study patients included 5 diabetics (2 insulin dependent; 3 non–insulin dependent). Triple-vessel coronary artery disease was present in 15 patients, double-vessel disease in 9, and single-vessel disease in 2 (coronary angiography was not available in 1 patient).
Regional Myocardial Blood Flow and Regional Rest Wall Motion
Under baseline conditions, the rate-pressure product at the initiation of the PET myocardial perfusion study (8269±2034 mm Hg/min) did not differ significantly from that at the beginning of the dobutamine stress echocardiogram (8197±1969 mm Hg/min). At rest, 350 segments (27 patients) with normal wall motion were available for analysis and had regional myocardial blood flow of 0.74±0.34 mL·g−1·min−1. There was no correlation between the rate-pressure product at rest and regional myocardial blood flow when all segments were considered. However, in a subset (n=30, 8 patients) with regional myocardial blood flow with adenosine >2.00 mL·g−1·min−1 (ie, segments capable of a dynamic range of flows), a statistically significant correlation was observed (r=.76, P<.01) between rate-pressure product at rest and rest regional myocardial blood flow.
Regional Myocardial Blood Flow and Wall Motion Response to High-Dose Dobutamine
The contractile response to high-dose dobutamine was divided into three categories: (1) normal augmentation (287 segments, 26 patients), (2) the development of hypokinesis (47 segments, 24 patients), and (3) the development of akinesis (16 segments, 7 patients). As shown in Table 2⇓, regional myocardial blood flow at rest in segments that augmented contraction normally in response to high-dose dobutamine (0.75±0.34 mL·g−1·min−1) did not differ from that of segments that became hypokinetic (0.74±0.36 mL·g−1·min−1). Segments that became akinetic at high-dose dobutamine had rest regional myocardial blood flow (0.59±0.24 mL·g−1·min−1) that did not differ significantly from that of the other groups.
In response to adenosine, regional myocardial blood flow increased equally in segments that augmented normally (1.06±0.72 mL·g−1·min−1) and in those that became hypokinetic (1.02±0.77 mL·g−1·min−1) at peak-dose dobutamine. In contrast, segments that became akinetic failed to increase myocardial blood flow with adenosine (0.68±0.30 mL·g−1·min−1). The myocardial flow reserve ratio (adenosine blood flow divided by rest blood flow) for segments that contracted normally at high-dose dobutamine (1.42±0.64) did not differ significantly from that of segments that became hypokinetic (1.48±1.03) or akinetic (1.21±0.49).
Similarly, percent diameter stenosis of vessels supplying myocardial segments that contracted normally at high-dose dobutamine (82±17%) did not differ significantly from that of segments that became hypokinetic (88±10%) or akinetic (92±8%). The percentage of segments supplied by a coronary vessel with no stenosis did not differ significantly between those that contracted normally (9%) at maximally tolerated dobutamine and those that became abnormal (7%; z=−0.49; P=NS). Myocardial flow reserve of segments perfused by coronary vessels without stenosis was 2.07±0.82 (n=28 segments of 7 patients). Finally, for all segments taken together (save those distal to a fully occluded vessel, which were not considered), a weak, albeit statistically significant, correlation was observed between myocardial blood flow with adenosine and percent diameter coronary artery stenosis (r=.50; P<.001; r2=.25).
Myocardial Flow Reserve in Normal Volunteers
Rest and adenosine-stimulated regional myocardial blood flows in nine normal volunteers (five men, four women; age, 44±13 years) studied in our laboratory under conditions identical to those of the present study were 1.00±0.25 and 3.24±0.87, respectively, with flow reserve ratio of 3.45±1.27.
Heart rate (beats per minute) and systolic arterial pressure (mm Hg) in volunteers were 67±9 and 133±17, respectively, at rest and 103±32 and 131±15 with adenosine. Systolic arterial pressure both at rest and with adenosine in volunteers was very similar to that of patients (rest, 133±17 and adenosine, 125±21) reported in this study. Heart rate at rest in patients (62±12) also was comparable to that of volunteers but was lower with adenosine (68±17) versus volunteers (103±32).
A patient-based analysis also was performed in which individuals on (n=15) and off (n=12) β-blockers were considered separately. Within each group, segments with normal and abnormal (hypokinesis or akinesis) wall motion response to high-dose dobutamine were compared in terms of rest and adenosine-stimulated regional myocardial blood flow. For each patient, myocardial blood flows of all segments with normal response to high-dose dobutamine were averaged for rest and adenosine conditions to reduce statistical noise. Likewise, blood flows at rest and with adenosine of all segments with abnormal response at high-dose dobutamine were averaged. As expected, rate-pressure product both at rest (9164±2111) and at high-dose dobutamine (17 490±5111) in the group off β-blockers exceeded (P<.05) that of patients on β-blockers (7552±1717 and 13 516±4037, respectively).
Myocardial segments of patients off β-blockers that contracted normally at high-dose dobutamine were able to increase rest myocardial blood flow (0.70±0.31 mL·g−1·min−1) in response to adenosine (0.86±0.46 mL·g−1·min−1, P<.05) although the absolute increment (≈0.17 mL·g−1·min−1) was small. In contrast, segments that failed to contract normally at high-dose dobutamine were unable to augment rest blood flow (0.63±0.24 mL·g−1·min−1) in response to adenosine (0.65±0.19 mL·g−1·min−1).
In patients on β-blockers, myocardial segments that contracted normally at high-dose dobutamine also demonstrated an increase in rest myocardial blood flow (0.78±0.31 mL·g−1·min−1) with adenosine (1.10±0.70 mL·g−1·min−1, P<.05). Again, the increment (≈0.32 mL·g−1·min−1) was not large. Myocardial segments that became abnormal at high-dose dobutamine, however, exhibited similar levels of blood flow both at rest (0.74±0.34 mL·g−1·min−1) and with adenosine (1.06±0.82 mL·g−1·min−1). Although the absolute increment was identical (≈0.32 mL·g−1·min−1) to that observed in segments that remained normal with high-dose dobutamine, the change failed to attain statistical significance because of smaller sample size and greater variation. Segments with abnormal wall motion at high-dose dobutamine, however, were much more likely to have been adjacent to a segment with abnormal wall motion at rest than segments that contracted normally with stress (48% versus 16%; χ2=16.1; P<.001). The same was true of myocardial segments in patients not on β-blockers (51% versus 21%; χ2=10.9; P<.01).
It also should be noted that 11 of 15 patients on β-blockers had a majority of segments that contracted normally at high-dose dobutamine but a minority that did not. These patients, therefore, exhibited both normal and abnormal wall motion responses that obviously could not be accounted for by differences either in rate-pressure product or β-blocker status, since in a given individual, all segments saw the same rate-pressure product and all were exposed to β-blocker. Further, rate-pressure product at high-dose dobutamine of the 4 patients who had all normal contractile responses (13 404±4717) was essentially the same as that of patients who exhibited mixed responses (13 557±4018). Finally, percent stenosis of coronary vessels that supplied segments of patients on β-blockers (84±17%) did not differ significantly from that of segments in patients off β-blockers (82±16%).
Dobutamine Echocardiography Sensitivity Analysis
Two patients could not be analyzed, one because coronary arteriograms were not available, the other because the only segments with normal wall motion at rest were in vascular territories without significant stenosis. Thus, of 25 patients remaining, 22 (84%) were identified by dobutamine stress echocardiography as having coronary artery disease. A total of 63 coronary arteries had ≥70% stenosis, of which 46 (73%) were identified by reversible wall motion abnormality of at least one segment in its vascular territory. It should be noted that 5 of 32 false-negative vessels (16%) were observed in patients off β-blockers versus 12 of 31 (39%; z=2.06; P<.05) in patients on β-blockers. Sensitivity for detection of abnormal vessels, therefore, was higher in patients off β-blockers (83%) in comparison with those on the drug (61%).
Analysis of FDG–Myocardial Blood Flow Relationship
Myocardial metabolic status, as assessed by FDG–blood flow match or mismatch, was analyzed on a segment basis (Table 3⇓). Those segments that augmented contraction normally with high-dose dobutamine had evidence of mismatch in only 63 of 269 segments (23%), whereas those that became abnormal (hypokinetic or akinetic) had mismatch in 24 of 59 segments (41%) (χ2=7.40; P<.01 [Fisher's exact test]). It should be noted that in 2 patients, 22 segments could not be analyzed, in one case because the scan could not be obtained and in the other because 4 segments were not adequately visualized. Thus, only 328 segments (269+59) were available for analysis instead of 350. Finally, it should be noted that 21 of 25 patients had at least 1 myocardial segment with FDG–blood flow mismatch.
Logistic Multiple Regression Analysis
To determine which variables were the best predictors of a myocardial contractile response to high-dose dobutamine stress, a logistic multiple regression analysis was performed. Independent variables included in the analysis were regional myocardial blood flow at rest and with adenosine; myocardial flow reserve ratio; delta regional myocardial blood flow, defined as regional myocardial blood flow with adenosine minus regional myocardial blood flow at rest ≥0.25 mL·g−1·min−1; rate-pressure product at high-dose dobutamine; β-blocker status; FDG match/mismatch status; and tethering factor (segment adjacent to one with abnormal wall motion at rest). Only variables that were statistically significant univariate predictors were included in the final model.
Tethering factor (odds ratio, 4.57; P<.001), β-blocker (odds ratio, 2.68; P<.01), and FDG match/mismatch status (odds ratio, 2.10; P<.05), respectively, were the strongest predictors of the regional wall motion response to high-dose dobutamine. The overall value of χ2 for the model was 45.13 (P<.0001). Although delta regional myocardial blood flow was a significant univariate predictor of wall motion response to dobutamine (odds ratio, 2.02; P<.02), it was excluded because of interaction effects with other variables in the model. Other myocardial blood flow parameters failed to emerge as univariate predictors of regional wall motion response to high-dose dobutamine.
Principal Findings and Physiological Considerations
This study tested the hypothesis in patients with stable ischemic heart disease that the ability of a myocardial segment to augment contraction normally in response to maximally tolerated inotropic stimulation with dobutamine is only partially explained by the ability of that segment to increase blood flow in response to a potent vasodilator stimulus such as adenosine. This hypothesis reflects the knowledge that the relationship between augmentation of regional myocardial contraction on the one hand and blood flow on the other is likely to be complex. Results of animal studies in fact indicate, in the face of a moderate to severe coronary artery stenosis and limited flow reserve, that regional contractile response to inotropic stimulation may be converted from abnormal to normal (or nearly so) by addition of β-blockade.4 5 9 Similarly, results of other animal studies indicate that (1) regional wall motion may become abnormal despite normal perfusion of the epicardium if endocardial perfusion and contraction is impaired21 and (2) depending on inotropic properties of the coronary vasodilator used, comparable degrees of coronary “steal” in the setting of stenosis may produce dissimilar degrees of myocardial ischemia.10 Accordingly, a substantial body of experimental data indicates that a simple, straightforward relationship between coronary flow reserve and segmental contractile response to inotropic stimulation does not always exist. The data obtained in the present study demonstrate that similar considerations apply in humans with stable ischemic heart disease.
In the present study, we found in patients not on β-blockers that the ability of myocardial segments to respond normally to maximally tolerated inotropic stimulation was associated with the ability of those segments to increase blood flow in response to adenosine. The absolute increment in blood flow achievable with adenosine in these segments, although very modest (≈0.2 mL·g−1·min−1), contrasted with that of myocardial segments that became abnormal with dobutamine and were unable to increase blood flow at all with adenosine. In patients taking β-blockers, myocardial flow reserve of segments that contracted normally at high-dose dobutamine also was limited and permitted only a modest increase in myocardial blood flow (≈0.3 mL·g−1·min−1) in response to adenosine. The increment, however, was greater than that observed in segments of patients not on β-blockers that also contracted normally at high-dose dobutamine and further indicates that the degree to which myocardial blood flow must increase to meet the demand of maximally tolerated dobutamine stress is not great.
These observations indicate that other factors may be equally or perhaps even more important than myocardial flow reserve in determining contractile response to dobutamine stress. Thus, myocardial segments of patients on β-blockers that failed to contract normally at high-dose dobutamine had levels of rest and adenosine-stimulated myocardial blood flow that were essentially the same as those of segments that contracted normally. The segments that became abnormal with stress, however, differed in two respects. First, these segments were much more likely to be adjacent to segments with abnormal wall motion at rest and thus more susceptible to the mechanical effect of tethering.21 Second, these segments also had a significantly higher frequency of FDG–blood flow mismatch and thus may have been more vulnerable to ischemic dysfunction,22 notwithstanding β-blocker therapy.
The physiological significance of FDG–myocardial blood flow mismatch in humans with stable ischemic heart disease is uncertain. Although acute myocardial ischemia results in enhanced glucose utilization and reduction in fatty acid metabolism by the myocardium,23 24 there are no data to indicate that such a state can be maintained more or less indefinitely. Moreover, results of animal studies indicate that the myocardium has substantial ability to adapt to acute ischemia.25 26 The fact that segments considered in this analysis all had normal wall motion at rest further argues against the notion that FDG–blood flow mismatch is indicative of ischemia at rest in humans with chronic stable ischemic heart disease. Nevertheless, FDG–blood flow mismatch probably is a marker of altered glucose metabolism by the myocardium22 27 and appears to be associated with increased vulnerability to ischemia, especially under conditions of catecholamine stimulation, with its known oxygen-wasting effect.28
As noted earlier, the observation that β-blocker therapy had a protective effect in terms of myocardial ischemia was anticipated.4 5 9 However, it should be recalled that in this study, a normal response to dobutamine was defined as an appropriate increase in regional contractile function, which, at least in theory, could have been inhibited rather than facilitated by β-blocking drugs. The most likely explanation for this apparent paradox is that the inotropic, chronotropic, and associated oxygen-wasting properties of dobutamine28 were sufficiently modulated by β-blockade to prevent ischemia but were not so completely blunted as to prevent an increase in regional contractility.
It is worth considering briefly the potential role of coronary stenosis severity and rate-pressure product at maximally tolerated dobutamine in accounting for the results of the study. Angiographic stenosis severity was very similar across the spectrum of regional wall motion responses and did not differ significantly among them. Accordingly, the argument that segments with normal wall motion response at high-dose dobutamine were supplied by coronary vessels that had little or no stenosis and those that exhibited abnormal wall motion were supplied by vessels with severe stenosis is directly refuted by the data. Furthermore, it should be stressed that segments that were hypokinetic at high-dose dobutamine were able to augment myocardial blood flow with adenosine to the same extent as those that contracted normally with maximally tolerated dobutamine. Thus, limitation on maximal myocardial blood flow, the true parameter of interest, for which stenosis severity serves as a weak proxy,29 is not sufficient to explain the results of the study. Similarly, as demonstrated by logistic multiple regression analysis, once tethering, β-blocker treatment, and metabolic (ie, FDG) status had been accounted for, rate-pressure product failed to make an independent contribution to predicting regional wall motion response to maximally tolerated dobutamine. Indeed, rate-pressure product failed to emerge even as a univariate predictor of regional wall motion at high-dose dobutamine. This is perhaps not surprising, since rate-pressure product does not reflect contractility and oxygen-wasting effects of dobutamine. In contrast, β-blocker therapy modulates both, in addition to lowering rate-pressure product, and so proved to be a better predictor. Moreover, rate-pressure product cannot account for either tethering or metabolic status, both of which proved to be very strong correlates of the regional wall motion response at maximally tolerated dobutamine.
Comparison With Other Reports
Several reports in the literature bear directly on the present study and should be considered. Perhaps of greatest interest are prior studies indicating that dobutamine stress echocardiography has both high sensitivity and high specificity for the diagnosis of coronary artery disease.19 20 In one report,20 a new or worsened wall motion abnormality was considered a positive dobutamine stress result, with the gold standard being a qualitatively scored coronary angiogram (>50% diameter reduction). This analysis was patient-based for purposes of determining sensitivity and specificity for diagnosis of ischemic heart disease and thus did not specifically address the issue of whether or not each segment in the vascular territory of a “significant” stenosis had a wall motion abnormality with dobutamine. As long as one segment was positive, the patient would be detected and a true-positive test response recorded. Analysis of the data in the present study in this fashion yielded a patient sensitivity of 84%.
In the other study,19 the primary analysis was based on detection of coronary stenosis (>50% diameter reduction by hand tracing and measurement of coronary arteriograms) in the three major vascular territories. These authors also made use of criteria that permitted persistent (rest versus stress) wall motion abnormality to count as a positive response. Key to this analysis, however, was the scoring method used and the assumption underlying it. Thus, the assumption was made that all stenoses with lesions >50% should be associated with wall motion abnormalities. (The authors' reported results were essentially unchanged when the data were reanalyzed with a criterion of ≥70% stenosis.) Since the echocardiogram was divided into multiple segments, each coronary artery in effect had multiple chances to be “positive.” A wall motion abnormality in only one of the segments subtended by a given vessel would suffice to score it positive. Potential discrepancies between segments in a single vascular distribution, some of which did and others of which did not exhibit a contraction abnormality with dobutamine, would not emerge from a vessel-based analysis and thus may have been overlooked. Analysis of the data obtained in the present study in similar fashion would yield an overall sensitivity for detection of ≥70% stenosis of individual coronary arteries of 73% (61% in patients on β-blockers and 83% in those off them).
It also should be emphasized that the present study considered only segments with normal wall motion at rest and correlated every segmental response to dobutamine with a PET measurement of myocardial blood flow and flow reserve. Furthermore, not only was the wall motion response of each segment to dobutamine considered, but also no assumptions were made about the physiological significance of any stenosis. The validity of the latter approach is nicely illustrated by the work of Picano et al,30 who demonstrated with quantitative coronary arteriography that in a population of patients with single-vessel coronary artery disease, 7 of 20 vessels with >50% stenosis failed to be associated with regional wall motion abnormality during dipyridamole echocardiography. Myocardial blood flow and flow reserve were assessed by PET/[13N]ammonia technique in these patients. While those with a positive study (n=11) on average had anatomically more severe stenoses and lower flow reserve, there was considerable functional overlap between the two groups in terms of maximal myocardial blood flow and flow reserve (R2=.41 for correlation between flow reserve and anatomic stenosis severity), an observation consistent with other data reported in the literature.31 32
More important is the fact that all 20 vessels would have qualified as “significant” for the purposes of the dobutamine echocardiographic studies19 20 enumerated above, notwithstanding the fact that their functional significance was variable. As long as the analysis of dobutamine (or dipyridamole) echocardiography stress tests is directed toward detection either of patients or of vessels and the assumption is made that every stenosis >50% (or >70%) is “significant,” then both sensitivity and specificity will be enhanced. This is particularly so in a vessel-based analysis in which only one of many potential segments need be positive for a true-positive result to be scored and any stenosis >50% (some of which will not cause functional abnormalities) is, ipso facto, deemed “significant.” As noted above, the data obtained in the present study, when analyzed on a patient or vessel basis, yielded sensitivity results for dobutamine echocardiography comparable to those reported by others.19 20
This study has certain limitations that should be considered. First, PET measurements of regional myocardial blood flow have been validated by our laboratory15 as well as others.33 34 Nevertheless, measurement of K1 is subject to a statistical error that was estimated in an earlier report15 to be on the order of ±0.10 mL·g−1·min−1 (1 SD). Second, echocardiographic assessment of regional wall motion was qualitative in nature. Although this is state of the art for clinical practice, wall motion assessment by quantitative means would be more precise. Nevertheless, previous work has demonstrated the reliability and reproducibility of this method.35
Third, while it would have been of interest to measure regional myocardial blood flow during dobutamine infusion, it was not possible to do so for this study. Nevertheless, the blood flow response of a segment to adenosine is widely accepted as indicative of its maximal flow potential. Segments unable to increase flow with adenosine, therefore, would not be expected to increase flow with dobutamine either. It is known, for instance, that dipyridamole, which is equally or less potent than adenosine, is a more potent coronary dilator than dobutamine.36 37
Furthermore, the effect of dobutamine on the coronary circulation in the presence of a hemodynamically important stenosis appears to depend on a variety of experimental factors, including species, dose, and associated hemodynamics (eg, heart rate and arterial pressure). Although one group has reported a decline in endocardial flow distal to a severe coronary stenosis,37 others have shown an increase.36 38 In the present study, the level of rate-pressure product at high-dose dobutamine was moderate (≈13 000 in patients on β-blockers and ≈19 000 in those off) and roughly double the respective levels at rest. Accordingly, in myocardial segments capable of increasing blood flow as demonstrated with adenosine, it is likely that blood flow also increased during dobutamine. The fact that segments that became akinetic with dobutamine in the present study also were unable to increase blood flow in response to adenosine supports this view. In contrast, in segments that were able to increase their flow to levels ≥2.0 mL·g−1·min−1 in response to adenosine, only 5 of 31 segments (16%; two patients) failed to increase contraction appropriately in response to dobutamine. In one of the patients, both segments were adjacent to ones with abnormal wall motion at rest and thus most likely reflect the effects of tethering. Abnormal segmental wall motion response in the other patient notwithstanding, excellent maximal myocardial blood flow and flow reserve may reflect other factors, such as qualitative scoring of regional wall motion.
Finally, because measurements of regional wall motion and regional myocardial blood flow were made by different techniques, the accuracy of registering myocardial segments should be considered. While every effort was made to ensure the fidelity of PET and echocardiographic description of similar regions, misregistration is possible and may have introduced some of the statistical noise that is present in the data. On balance, however, these errors probably were not large, given the strength of the statistical associations observed and the complete separation of the processes used in grading regional wall motion and in measuring regional myocardial blood flow.
In summary, this study demonstrates in patients with stable ischemic heart disease that the ability of a myocardial segment to respond normally to maximally tolerated inotropic stimulation with dobutamine is only partially dependent on the ability of the segment to increase blood flow in response a maximal vasodilator stimulus, such as adenosine. The ability to increase myocardial blood flow on average is necessary but not sufficient for a normal contractile response to maximally tolerated dobutamine stress. Mechanical forces related to tethering, β-blocker therapy, and myocardial metabolic status also play an important role.
Further, the data obtained demonstrate that the absolute increment in myocardial blood flow required to meet the demand of maximally tolerated inotropic stimulation with dobutamine is not likely to be large even in patients who are not on β-blocker therapy (average, ≈0.2 mL·g−1·min−1). Thus, development of abnormal wall motion during high-dose dobutamine implies either an inability of that segment to increase myocardial blood flow at all or else the presence of another factor, such as tethering or metabolic abnormality, as reflected by FDG–blood flow mismatch.
Finally, the data strongly indicate that even though the sensitivity of dobutamine stress echocardiography for detection of stenosed coronary vessels may be high, many myocardial segments with limited flow reserve nevertheless will have normal wall motion at maximally tolerated dobutamine, even in patients who are not on β-blockers. Of segments with normal contraction at rest in patients off β-blockers, 70% had normal contraction at high-dose dobutamine in the present study. Clinically, the choice of imaging study in evaluating patients for known or suspected ischemic heart disease should be made with these considerations in mind.
Edie Sinagra assisted with preparation of the manuscript and coordination of patient scheduling. We wish to express our appreciation to the technical personnel of the PET, echocardiography, and nuclear cardiology laboratories for dedicated and skilled assistance in the performance of these studies. Jimmy Efird, PhD, performed the logistic regression analysis.
- Received November 8, 1995.
- Revision received February 5, 1996.
- Accepted February 20, 1996.
- Copyright © 1996 by American Heart Association
Canty JMJ, Klocke FJ. Reductions in regional myocardial function at rest in conscious dogs with chronically reduced regional coronary artery pressure. Circ Res. 1987;61(suppl II):II-107-II-116.
Vatner S. Correlation between acute reductions in myocardial blood flow and function in conscious dogs. Circ Res. 1980;47:201-207.
Bache RJ, Schwartz JS. Myocardial blood flow during exercise after gradual coronary occlusion in the dog. Am J Physiol. 1983;245(Heart Circ Physiol 14):H131-H138.
Kumada T, Gallagher KP, Shirato K, McKown D, Miller M, Kemper WS, White F, Ross J. Reduction of exercise induced regional myocardial dysfunction by propranolol: studies in a canine model of chronic coronary artery stenosis. Circ Res. 1980;46:190-200.
Suga H, Hisano R, Goto Y, Yamada O, Igarashi Y. Effect of positive inotropic agents on the relation between oxygen consumption and systolic pressure volume area in canine left ventricle. Circ Res. 1983;53:306-318.
Gallagher KP, Matsuzaki M, Osakada G, Kemper WS, Ross JR Jr. Effect of exercise on the relationship between myocardial blood flow and systolic wall thickening in dogs with acute coronary stenosis. Circ Res. 1983;52:716-729.
Gewirtz H, Gross SL, Williams DO, Most AS. Contrasting effects of nifedipine and adenosine on regional flow distribution and metabolism distal to a severe coronary arterial stenosis: observations in sedated, closed-chest domestic swine. Circulation. 1984;69:1048-1057.
Ross JR Jr. Myocardial perfusion-contraction matching: implications for coronary heart disease and hibernation. Circulation. 1991;83:1076-1083.
Bolli R. Myocardial ‘stunning’ in man. Circulation. 1992;86:1671-1691.
Gewirtz H, Fischman AJ, Abraham SA, Gilson M, Strauss HW, Alpert NM. Positron emission tomographic measurements of absolute regional myocardial blood flow permits identification of nonviable myocardium in patients with chronic myocardial infarction. J Am Coll Cardiol. 1994;23:851-859.
Rosenspire KC, Schwaiger M, Magner TJ, Hutchins GD, Sutorik A, Kuhl DE. Metabolic fate of 13-N-ammonia in human and canine blood. J Nucl Med. 1990;31:163-167.
Gewirtz H, Fischman A, Alpert N. Validation of K1 parametric images obtained from dynamic 13-NH3 PET images: comparison to K1 values obtained with a standard non-linear estimation program. J Nucl Med. 1994;35:77P. Abstract.
Marwick T, Willemart B, D'Hondt A, Baudhuin T, Wijns W, Detry J, Melin J. Selection of the optimal nonexercise stress for the evaluation of ischemic regional myocardial dysfunction and malperfusion: comparison of dobutamine and adenosine using echocardiography and 99mTc-MIBI single photon emission computed tomography. Circulation. 1993;87:345-354.
Forster T, McNeill A, Salustri A, Reijs A, el-Said E, Roelandt J, Fioretti P. Simultaneous dobutamine stress echocardiography and technetium-99m isonitrile single-photon emission computed tomography in patients with suspected coronary artery disease. J Am Coll Cardiol. 1993;21:1591-1596.
Gallagher KP, Osakada G, Hess OM, Koziol JA, Kemper WS, Ross JRJ. Subepicardial segmental function during coronary stenosis and the role of myocardial fiber orientation. Circ Res. 1982;50:352-359.
Buxton DB, Mody FV, Krivokapich J, Phelps ME, Schelbert HR. Quantitative assessment of prolonged metabolic abnormalities in reperfused canine myocardium. Circulation. 1992;85:1842-1856.
Mjos OD, Kjekshus JK, Lekven J. Importance of free fatty acids as a determinant of myocardial oxygen consumption and myocardial ischemic injury during norepinephrine infusion in dogs. J Clin Invest. 1974;53:1290-1299.
Fedele FA, Gewirtz H, Capone RJ, Sharaf B, Most AS. Metabolic response to prolonged reduction of myocardial blood flow distal to a severe coronary artery stenosis. Circulation. 1988;78:729-735.
Pantley GA, Malone SA, Rhen WS, Anselone CG, Arai A, Bristow J, Bristow JD. Regeneration of myocardial phosphocreatine in pigs despite continued moderate ischemia. Circ Res. 1990;67:1481-1493.
Liedtke J, Renstrom B, Nellis S. Correlation between [5-3H]glucose and [U-14C]deoxyglucose as markers of glycolysis in reperfused myocardium. Circ Res. 1992;71:689-700.
Vanoverschelde JLJ, Wijns W, Essamri B, Bol A, Robert A, Labar D, Cogneau M, Michel C, Melin JA. Hemodynamic and mechanical determinants of myocardial O2 consumption in normal human heart: effects of dobutamine. Am J Physiol. 1993;265(Heart Circ Physiol 34):H1884-H1892.
Topol EJ, Nissen SE. Our preoccupation with coronary luminology: the dissociation between clinical and angiographic findings in ischemic heart disease. Circulation. 1995;92:2333-2342.
Picano E, Parodi O, Lattanzi F, Sambuceti G, Andrade MJ, Marzullo P, Giorgetti A, Salvadori P, Marzilli M, Distante A. Assessment of anatomic and physiological severity of single-vessel coronary artery lesions by dipyridamole echocardiography: comparison with positron emission tomography and quantitative arteriography. Circulation. 1994;89:753-761.
Demer LL, Gould KL, Goldstein RA, Kirkeeide RL, Mullani NA, Smalling RW, Nishikawa A, Merhige ME. Assessment of coronary artery disease severity by positron emission tomography: comparison with quantitative coronary arteriography in 193 patients. Circulation. 1989;79:825-835.
Di Carli M, Czernin J, Hoh C, Gerbaudo V, Brunken R, Huang S, Phelps M, Schelbert H. Relation among stenosis severity, myocardial blood flow, and flow reserve in patients with coronary artery disease. Circulation. 1995;91:1944-1951.
Krivokapich J, Smith GT, Huang SC. 13-N ammonia myocardial imaging at rest and with exercise in normal volunteers: quantification of absolute myocardial perfusion with dynamic positron emission tomography. Circulation. 1989;80:1328-1337.
Schnittger I, Fitzgerald P, Gordon E, Alderman E, Popp R. Computerized quantitative analysis of left ventricular wall motion by two-dimensional echocardiography. Circulation. 1984;70:242-254.
Fung AY, Gallagher KP, Buda AJ. The physiological basis of dobutamine as compared with dipyridamole stress interventions in the assessment of critical coronary stenosis. Circulation. 1987;76:943-951.
Zhang J, Path G, Chepuri V, Homans DC, Merkle H, Hendrich K, Ugurbil K, Bache RJ, From AHL. Effects of dobutamine on myocardial blood flow, contractile function, and bioenergetic responses distal to coronary stenosis: implications with regard to dobutamine stress testing. Am Heart J. 1995;129:330-342.