Remote Noninfarcted Region Dysfunction Soon After First Anterior Myocardial Infarction
A Magnetic Resonance Tagging Study
Background Previous studies have demonstrated hyperkinetic endocardial motion of noninfarcted myocardium early after myocardial infarction (MI). We wished to substantiate the findings of increased function of remote noninfarcted regions using magnetic resonance (MR) myocardial tagging in patients soon after anterior MI.
Methods and Results Twenty-eight patients (25 male; mean age, 52 years) were studied on day 5±2 after first anterior MI. All had single-vessel left anterior descending coronary artery (LAD) disease and had received reperfusion therapy but had evidence of regional left ventricular (LV) dysfunction and an ejection fraction (EF) ≤50%. Breath-hold, segmented k-space, gradient-echo MR tagging was performed with short-axis imaging spanning the LV. Percent circumferential shortening (%S) on a topographic basis, LV mass, and EF were measured. Regional %S was compared with that in 10 normal subjects (7 male; mean age, 43 years). We found reduced intramyocardial %S throughout the LV in the patient group. Percent shortening was lower in patients compared with control subjects at all sites along the long axis of the ventricle (9±5% versus 23±3% at the apex, P<.0001; 11±5% versus 21±3% at the midventricle, P<.0001; 14±3% versus 17±5% at the base, P<.02). The basal lateral and midinferior regions, remote from LAD territory, demonstrated reduced %S and a strong trend toward reduced %S, respectively.
Conclusions Patients on day 5 after first anterior MI with single-vessel disease demonstrate reduced intramyocardial circumferential shortening throughout the LV, including remote noninfarcted regions. Potential mechanisms include altered coronary vasodilatory properties, changes in regional mechanical load, or mechanical tethering to infarcted regions.
Early after coronary occlusion, regional function within remote noninfarcted myocardium depends, in part, on the length of time after the ischemic insult. Dysfunction in remote myocardium has been demonstrated as early as 90 seconds1 and at 30 minutes2 after coronary occlusion. In patients within 12 hours of MI, previous qualitative echocardiographic studies have suggested normal or hyperdynamic function in remote noninfarcted areas in many patients.3
Methods developed to quantitatively assess regional myocardial function early after MI in humans include contrast ventriculography,4 echocardiographic endocardial mapping,5 and ultrafast cine CT.6 The recent development7 8 and validation9 10 of MR myocardial tagging have permitted the analysis of intramyocardial segmental function in normal subjects11 and patients with myocardial disease.12 13
With the advent of breath-hold MR tissue tagging,14 15 a comprehensive quantitative evaluation of regional LV function in humans by MRI soon after MI can be obtained in ≤30 minutes. We used these rapid MRI techniques to evaluate regional intramyocardial function throughout the LV in patients early after reperfused anterior infarction with single-vessel disease of the LAD to determine whether remote noninfarcted myocardium is hyperkinetic relative to normal subjects.
This study was approved by the Institutional Review Board of Allegheny General Hospital, and all human subjects gave informed consent. Ten normal subjects, including 7 men and 3 women 43±14 years old, underwent MRI. None had clinical or echocardiographic evidence of cardiac disease.
Twenty-eight patients (25 men and 3 women, 52±12 years old) were studied with MRI on day 5±2 after anterior MI. The diagnosis of MI was made in the conventional manner from clinical history, ECG, and plasma creatine kinase levels elevated to more than twice the normal level with MB fraction >5%. Their clinical characteristics are listed in Table 1⇓.
All patients had single-vessel disease involving the LAD without significant (≥50%) disease in the other major epicardial coronary arteries. Four had been treated with TPA, 16 with primary angioplasty, and 8 with TPA followed by rescue angioplasty. Data on time to therapy were not available in 4 patients. The average time from onset of chest pain to initiation of reperfusion therapy was 288±151 minutes in the remaining 24 patients. The peak creatine kinase was 2678±2066 U/L. All had evidence of regional LV dysfunction and an EF ≤50% by left ventriculography or echocardiography before entry into the study. The mean EF by these imaging techniques was 40±8%. All had documented patent infarct-related arteries by angiography.
During the MR scan, the mean heart rate was 79±13 bpm, systolic blood pressure 112±15 mm Hg, and diastolic blood pressure 68±8 mm Hg. The patients' medications at the time of the imaging session included aspirin in 26, ACE inhibitors in 18, β-blockers in 17, nitrates in 15, warfarin in 6, digoxin in 3, diuretics in 2, persantine in 2, and a calcium channel blocker in 1.
Magnetic Resonance Imaging
MRI was performed in a Siemens Magnetom SP 1.5-T scanner with the patient prone on an elliptical spine surface coil. Scout images were obtained to localize the long axis of the LV, followed by a short-axis cine series in a plane near the mitral valve that provided images every 40 ms to identify end systole as the point of minimum LV cavity volume. A series of multiphase short-axis single-slice tagged images were obtained by use of a breath-hold multiple phase–encoded gradient-echo method and segmented k-space trajectory. The tag pattern was created with a 1331 binomial radiofrequency pulse series adjusted to produce tags every 7 mm in a diagonal orientation. The field of view was 280 mm and matrix size 128×256, interpolated to 256×256 for display, yielding a pixel size of 1.09×1.09 mm. Image planes 7 mm thick were generated, spanning the entire LV cavity from base to apex, which required an average of 12 slices per patient (Fig 1⇓). The breath-hold spanned 18 heartbeats, or ≈14 seconds per breath-hold at the average heart rate of the patients in the study. TR (repetition time) was adjusted (from a minimum of 58 ms to a maximum of 90 ms) so as to time one image during the five-phase image series at end systole. The entire imaging session lasted ≈45 minutes.
Quantitative analysis of images was performed by a single investigator (T.M.T.) using an operator-driven image analysis tool developed in the Volumetric Image Display and Analysis software package (VIDA, University of Iowa) loaded on a Sun workstation. LV mass was calculated from planimetered epicardial and endocardial areas of interleaved short-axis end-diastolic tagged images. LV mass was calculated with a modified Simpson's rule according to the following formula16 : LV mass=Σ(epi−endo)(7)(1.19)(0.001)(1.05), where mass is expressed in grams, epi and endo represent the total area in pixels enclosed by the epicardium and endocardium, respectively, 7 is the slice thickness in millimeters, 1.19 is the area in square pixels in square millimeters, 0.001 was used to convert from cubic millimeters to cubic centimeters, and 1.05 represents the density of myocardium in grams per cubic centimeter. LVEF was calculated by use of end-diastolic and end-systolic volumes calculated from planimetered end-diastolic and end-systolic areas in the same fashion as LV mass, except that the density factor was deleted from the calculation.
Quantitative analysis of shortening was performed with an operator-driven interstripe distance–measuring software tool developed in VIDA and according to previously published methods. %S was measured at endocardial, midwall, and epicardial sites in the anterior, lateral, septal, and inferior regions in the LV short axis in all of the slices from apex to base in each study. The total number of slices along the long axis of the ventricle was divided into apical, midventricular, and basal thirds (four slices each on average), and data from these slices were averaged.
Comparisons of %S between patient and control groups were made with a two-tailed Student's t test for unpaired comparisons. Regions were compared individually by transmural site. In addition, data from all three transmural sites were averaged, and data from all regions around the short axis and from all three regions along the ventricular long axis were averaged. An unpaired t test was also used to compare results from patients with and without LVH. Univariate and stepwise multivariate linear regression analyses were performed to evaluate the effects of age, LV mass index, peak creatine kinase, and time to reperfusion therapy on %S. To examine for effects of type of reperfusion therapy (TPA, percutaneous transluminal coronary angioplasty, or both) or medications used at the time of the scan (β-blockers, nitrates, or ACE inhibitors) on %S in patients, ANOVA (factorial and three-way, respectively) was performed. Results are displayed as mean±SD. A value of P<.05 was considered to be significant.
Global LV Parameters
A total of 12±2 slices were imaged from base to apex in the patient group and 11±1 slices in the normal subjects. The mean EF in patients by MRI was 41±12%. The mean LV mass index was 112±20 g/m2. Eight of the patients had LVH by Framingham criteria,17 6 men and 2 women (patients 3, 4, 5, 11, 14, 17, 25, and 26; Table 1⇑). The mean LV mass index in the patients with LVH was 132±13 g/m2 compared with 102±14 g/m2 in those without LVH (P<.0001). The mean EF by MRI was 37±9% in the patients with LVH versus 43±12% in those without LVH (P=NS).
Intramyocardial Segment Shortening
%S was compared between patients and normal subjects by transmural location and region in both patients and normal subjects and is displayed in Table 2⇓. In the apex and midventricle, which included the infarcted region in patients with anteroapical MI, each region at each transmural location except the mid inferior wall demonstrates reduced %S. At the base, circumferential shortening was depressed in patients compared with normal subjects in the anterior and lateral subendocardium and the anterior midwall (Table 2⇓).
When data from the three transmural locations were averaged by region, only the values in patients in the basal septum and basal inferior walls were similar to those of normal subjects (Fig 2⇓ and Table 2⇑). The basal lateral wall, remote from anterior infarction and supplied by coronary arteries without significant disease, demonstrated reduced %S (17±5% versus 22±7% in normal subjects, P<.04). In the mid inferior wall, there was a strong trend toward lower %S (12±10% in patients compared with 19±5% in normal subjects, P<.08).
When data from all three transmural locations and four sites around the short axis were averaged, %S was lower in patients at all sites along the long axis of the ventricle, most markedly at the apex and midventricle, which included infarcted tissue (Fig 3⇓). At the apex, %S was 9±5% in patients versus 23±3% in normal subjects (P<.0001), at the midventricle 11±5% versus 21±3% (P<.0001), and at the base 14±3% versus 17±5% (P<.02). Within the anterior, septal, and lateral regions of patients, %S was lower than in normal subjects when values from three transmural sites and all slices from apex to base were averaged (Fig 4⇓). Within the anterior wall, %S was 7±5% compared with 17±4% in normal subjects (P<.0001). In the septum, %S in patients was 9±5% versus 20±4% in normal subjects (P<.0001) and in the lateral region, 16±6% compared with 26±4% (P<.0001). In the inferior wall, %S was 14±8% in patients and 18±4% in human volunteers (P=NS).
To determine whether preexistent LVH accounted for reduced intramyocardial function in remote noninfarcted regions, we compared %S by region in the patients with and without LVH. %S was not different between these two groups at any locus. At the base, %S was 13±5% in the 8 patients with LVH and 14±4% in those without (P=NS). By univariate regression analysis, there was a weak negative relationship between LV mass index and basal %S (P<.03, R2=.20). However, by multivariate regression analysis, no significant relationship was found between basal %S and age, LV mass index, peak creatine kinase, or time to reperfusion (n=24). By ANOVA, no significant effect of type of reperfusion therapy or medication used (β-blocker, ACE inhibitor, or nitrates) on %S was found in the patient group.
Using breath-hold MR myocardial tagging, we found reduced intramyocardial circumferential shortening throughout the LV in patients 1 week after anterior MI due to single-vessel disease of the LAD, including basal regions remote from the infarcted area. The largest reduction in function was, as expected, in the infarcted apical region. However, the basal lateral region, re-mote from LAD territory infarction and supplied by coronary arteries without significant disease, demonstrated reduced circumferential shortening compared with normal subjects. The mid inferior wall, similarly remote from the infarct, showed a strong trend toward reduced %S.
Hyperkinesis of remote noninfarcted myocardium may be present in the first hours after MI.3 18 19 20 21 This finding may be due to changes in regional afterload and local preload and not to effects of catecholamines.22 Semiquantitative analysis of echocardiograms at 12 hours after MI demonstrates hyperkinesis in up to 67% of patients.3 More quantitative measures using the centerline method show hyperkinesis in only 32% of patients in these early hours.23 Early remote-region hyperkinesis is less demonstrable in patients with multivessel coronary disease and may portend a worse prognosis.3 19
Later after infarction, the documentation of hyperkinesis in remote regions has been inconsistent. Theroux et al24 demonstrated an increase in %S from 20.7±0.8% to 27.2±2.4% (P<.05) at 1 week, due in part to an increase in the end-diastolic length of 5% in that segment. They hypothesized that eccentric hypertrophy might account for improved function in these regions. Angiographic studies have demonstrated that reperfusion has beneficial effects on function in remote regions in patients who undergo contrast ventriculography immediately and then at 10 days after reperfusion.25 These investigators found very mild hyperkinesis in noninfarcted regions at 10 days, more in the thrombolytic-treated patients than in the control subjects. The centerline method used in this study is sensitive to endocardial translation, which may reflect tethering effects and not intramyocardial function. Other ventriculographic studies have shown “regression toward the mean” at 1 week, with reduced function in originally hyperkinetic zones.21
Cine MRI has been used to evaluate LV wall motion in patients at a mean of day 7 after infarction.26 These infarcts were a mix of inferior and anterior infarcts, and the temporal resolution used in the imaging (eight phases per cardiac cycle) was limited. Of segments analyzed, 4% demonstrated >2 SDs above normal values for absolute or relative wall thickening. The quantitative analysis performed was a variation on the centerline method and is affected by through-plane motion.
Our previous animal studies after MI using MR tagging have not demonstrated remote-region hyperkinesis in the first week after MI. In an ovine model, we showed a downward trend in %S in remote noninfarcted tissue from before to 1 week after MI, from 20% to 18% in the subendocardium and from 16% to 12% in the subepicardium.16 In a two-dimensional analysis of mechanical function from MR tagging data in the same group of animals, the magnitudes of the principal strains in remote noninfarcted regions were not significantly different from before to 1 week after MI.27
Early dysfunction in regions remote from the infarct after anterior MI is unlikely to represent overt ischemia in the absence of diseased epicardial arteries supplying these regions. However, in a rat model, Karam et al28 showed that coronary flow and vasodilator reserve in noninfarcted myocardium were depressed. More recently, others have demonstrated reduced coronary blood flow reserve in remote noninfarcted regions in a patient population very similar to the one in this study.29 Altered vasodilator responsiveness in a region characterized by hyperkinesis earlier in the postinfarction period may induce a state of relative hypoperfusion, leading to later local mechanical dysfunction.20
Remote loading effects of infarcted tissue are a potential explanation for reduced circumferential shortening in remote regions. Changes in LV geometry after anterior infarction alters regional radii of curvature, increasing wall stresses in noninfarcted regions, as previously estimated in the chronic setting.30 Elevated systemic blood pressure as one component of load is not an explanation, because the mean pressure at the time of the study in the patient group was 112/68 mm Hg. Mechanical tethering of noninfarcted regions to infarcted myocardium could also be responsible for altered ventricular mechanics within noninfarcted regions.
Reduced %S could, in theory, be due to the presence of LVH. Using spin-echo MR tagging, Palmon et al12 demonstrated reduced %S in hypertensive LVH despite normal EF. Other techniques, including echocardiography, also have demonstrated reduced midwall shortening.31 However, there was no difference between %S in patients with and without LVH. By univariate regression analysis, basal %S was weakly related to LV mass index, although in a multivariate analysis no significant relationship was found.
Pharmacological effects could be associated with depressed shortening. All patients were taking agents that could affect regional circumferential shortening. However, no effect of β-blockers, ACE inhibitors, or nitrates on %S within the patient group was found. We have shown that ACE inhibition is associated with improved function in adjacent noninfarcted myocardium and has no significant effect in remote tissue in an ovine model of anteroapical infarction and LV remodeling.32 In addition, our group has shown no significant difference in %S in pharmacologically treated and untreated hypertensive subjects.12
The patient population studied is highly selected. Our findings may not be applicable to patients with inferior infarction, multivessel disease, preserved global LV function after reperfusion, or those who have not received reperfusion therapy. The infarcts are large, with an average EF of 40%, in part because of the relatively late time to reperfusion. A large subgroup received primary percutaneous transluminal coronary angioplasty in lieu of thrombolysis. The total number of patients studied is small, but the density of measurements permitted by MR tagging increased the power of our study to detect regional differences in intramyocardial function. The effect of medical therapy on regional function in the patient group is a consideration, because none of the control subjects were on medication. However, no systematic effect of medication on %S within the patient group was found.
In summary, breath-hold MR tagging was used to characterize regional intramyocardial function in a quantitative and topographic manner early after MI. Patients in the first week after anterior MI with single-vessel disease and LV dysfunction demonstrate reduced intramyocardial circumferential shortening throughout the LV, including remote regions supplied by coronary arteries without significant disease. Potential mechanisms include altered coronary vasodilatory properties, changes in regional mechanical load, and mechanical tethering effects. Further studies late after infarction in this patient group may improve the understanding of remote region dysfunction after anterior MI and its potential relationship to global LV remodeling.
This work was supported by NHLBI grant HL-42958 and a grant from the Allegheny-Singer Research Institute. The authors acknowledge the assistance of L. Dennis Wan, PhD, software engineer, and the magnetic resonance technologists at Allegheny General Hospital, especially June Yamrozik and Lois Miller.
Selected Abbreviations and Acronyms
|LAD||=||left anterior descending coronary artery|
|LV||=||left ventricle, left ventricular|
|LVH||=||left ventricular hypertrophy|
|MRI||=||magnetic resonance imaging|
|%S||=||percent intramyocardial circumferential segment shortening|
|TPA||=||tissue-type plasminogen activator|
- Received September 27, 1995.
- Revision received February 12, 1996.
- Accepted February 17, 1996.
- Copyright © 1996 by American Heart Association
Sheehan FH, Bolson EL, Dodge HT, Mathey DG, Schofer J, Woo HW. Advantages and applications of the centerline method for characterizing regional ventricular function. Circulation. 1986;74:293-305.
Picard MH, Wilkins GT, Ray PA, Weyman AE. Natural history of left ventricular size and function after acute myocardial infarction. Circulation. 1990;82:484-494.
Yeon SB, Reichek N, Palmon LC, Tallant B, Brownson D, Hoffman EA, Axel L. Validation of circumferential segment shortening by magnetic resonance tagging. Circulation. 1990;82:488. Abstract.
Clark N, Reichek N, Bergey P, Hoffman EA, Brownson D, Palmon L, Axel L. Normal segmental myocardial function: assessment by magnetic resonance imaging using spatial modulation of magnetization. Circulation. 1991;84:67-74.
Palmon L, Reichek N, Yeon SB, Clark NR, Brownson D, Hoffman EA, Axel L. Intramural myocardial shortening in hypertensive left ventricular hypertrophy with normal pump function. Circulation. 1994;89:122-131.
Kramer CM, Reichek NR, Ferrari VA, Theobald T, Dawson J, Axel L. Regional heterogeneity of function in hypertrophic cardiomyopathy. Circulation. 1994;90:186-194.
Fayad ZA, Axel L, Kraitchman DL. Magnetic resonance tagging: conventional spin-echo versus fast breath-hold gradient-echo imaging using phased-array coils. Proc Soc Magn Reson Med. 1993;3:1217. Abstract.
Kramer CM, Lima JAC, Reichek N, Ferrari VA, Llaneras MR, Palmon LC, Yeh I-T, Tallant B, Axel L. Regional function within noninfarcted myocardium during left ventricular remodeling. Circulation. 1993;88:1279-1288.
Levy D, Anderson KM, Savage DD, Kannel WB, Christiansen JC, Castelli WP. Echocardiographically detected left ventricular hypertrophy: prevalence and risk factors. Ann Intern Med. 1988;108:7-13.
Kerber RE, Marcus ML, Wilson R, Ehrhardt J, Abboud FM. Effects of acute coronary occlusion on the motion and perfusion of the normal and ischemic interventricular septum. Circulation. 1976;54:928-935.
Stamm RB, Gibson RS, Bishop HL, Carabello BA, Beller GA, Martin RP. Echocardiographic detection of infarct-localized asynergy and remote asynergy during acute myocardial infarction: correlation with the extent of angiographic coronary disease. Circulation. 1983;67:233-244.
Homans DC, Sublett E, Elsperger KJ, Schwarz JS, Bache RJ. Mechanisms of remote myocardial dysfunction during coronary artery occlusion in the presence of multivessel disease. Circulation. 1986;74:588-596.
Grines CL, Topol EJ, Califf RM, Stack RS, George BS, Kereiakes D, Boswick JM, Kline E, O'Neill WW, on behalf of the TAMI Study Group. Prognostic implications and predictors of enhanced regional wall motion of the noninfarct zone after thrombolysis and angioplasty therapy of acute myocardial infarction. Circulation. 1989;80:245-253.
Smalling RW, Ekas RD, Felli PR, Binion L, Desmond J. Reciprocal functional interaction of adjacent myocardial segments during regional ischemia: an intraventricular loading phenomenon affecting apparent regional contractile function in the intact heart. J Am Coll Cardiol. 1986;7:1335-1346.
Stadius ML, Maynard C, Fritz JK, Davis K, Ritchie JL, Sheehan F, Kennedy JW. Coronary anatomy and left ventricular function in the first 12 hours of acute myocardial infarction: the Western Washington Randomized Intracoronary Streptokinase Trial. Circulation. 1985;72:292-301.
Theroux P, Ross J Jr, Franklin D, Covell JW, Bloor CM, Sasayama S. Regional myocardial function and dimensions early and late after myocardial infarction in the unanesthetized dog. Circ Res. 1977;40:158-165.
Martin GV, Sheehan FH, Stadius M, Maynard C, Davis KB, Ritchie JL, Kennedy JW. Intravenous streptokinase for acute myocardial infarction: effects on global and regional systolic function. Circulation. 1988;78:258-266.
Lima JAC, Ferrari VA, Reichek N, Kramer CM, Palmon L, Llaneras MR, Tallant B, Young AA, Axel L. Segmental motion and deformation of transmurally infarcted myocardium in acute postinfarct period. Am J Physiol. 1995;268:H1304-H1312.
Karam R, Healy BP, Wicker P. Coronary reserve is depressed in postmyocardial infarction reactive cardiac hypertrophy. Circulation. 1990;81:238-246.