Combined Angiotensin II Receptor Antagonism and Angiotensin-Converting Enzyme Inhibition Further Attenuates Postinfarction Left Ventricular Remodeling
Background—ACE inhibition (ACEI) attenuates post–myocardial infarction (MI) LV remodeling, but the effects of angiotensin II type 1 receptor (AT1) antagonism alone or in combination with ACEI are unclear. Accordingly, we investigated the effects of AT1 antagonism, ACEI, and their combination in a well-characterized ovine postinfarction model.
Methods and Results—Beginning 2 days after transmural anteroapical MI, 62 sheep were treated with 1 of 5 treatment regimens: no therapy (control, n=12), standard-dose ACEI (sACEI; ramipril 10 mg/d, n=14), high-dose ACEI (hACEI; ramipril 20 mg/d, n=8), AT1 blockade (losartan 50 mg/d, n=13), and combination therapy with sACEI+AT1 blockade (CT; ramipril 10 mg/d+losartan 50 mg/d, n=15). MRI was performed before and 8 weeks after MI to quantify changes in LV end-diastolic and end-systolic volume indices (ΔEDVI, ΔESVI) and ejection fraction (ΔEF). Change in regional percent intramyocardial circumferential shortening in noninfarcted segments adjacent to the infarct (Adj Δ%S) was measured by tagged MRI. CT resulted in the most marked blunting of LV remodeling: ΔESVI (+1.0±0.4, +0.7±0.4, +0.6±0.3†, +0.9±0.5, and +0.4±0.2* mL/kg); ΔEDVI (+0.9±0.4, +0.7±0.5, +0.6±0.5, +0.9±0.5, and +0.4±0.3‡ mL/kg); ΔEF (−24±7, −18±6, −14±7†, −18±10, and −11±9* %); and Adj Δ%S (−8±4, −7±3, −5±3, −5±3, and −2±3* %) for Control, sACEI, hACEI, AT1 blockade, and CT, respectively (*P<0.04 versus sACEI, AT1 blockade, and control; †P<0.05 versus control; ‡P<0.002 versus AT1 blockade and control). EDVI and ESVI at 8 weeks after MI were smallest with CT (P<0.02 versus all).
Conclusions—Combination therapy with sACEI+AT1 blockade shows promise in attenuating postinfarction LV remodeling but was not clearly superior to hACEI in the present study.
Left ventricular (LV) remodeling after myocardial infarction is characterized by infarct expansion,1 2 myocyte hypertrophy,3 4 myocyte slippage,5 lengthening of noninfarcted segments,6 7 8 global ventricular dilatation,7 and dysfunction in noninfarcted regions adjacent to the infarct.8 ACE inhibition (ACEI) attenuates postinfarction LV remodeling in experimental models9 10 11 and improves survival in humans.12 Although ACEI blunts postinfarction LV remodeling, the mechanisms involved remain incompletely understood, and the degree of attenuation is incomplete.13 Experimental evidence indicates that the influence of ACEI on LV remodeling may involve both direct angiotensin II (Ang II) effects acting via a variety of Ang II receptor subtypes and indirect effects on kallikrein-kinin systems.14 15 Alternative (non-ACE) enzymatic pathways for the production of Ang II have also been demonstrated within the heart, which may limit the effectiveness of ACEI.16 16A Combined ACEI and Ang II type 1 receptor (AT1) blockade has also been shown to provide enhanced benefits in a pacing model of congestive heart failure.17 The addition of AT1 blockade to recommended or maximally tolerated ACEI was recently shown to improve exercise capacity and reduce symptoms in patients with heart failure.18 Jorde et al19 recently found, however, that maximally recommended doses of ACEI may be inadequate and that very-high-dose ACEI provided experimental hemodynamic effects similar to those with combined AT1 blockade and standard-dose ACEI in patients with heart failure. The effects of combined ACEI and AT1 antagonism on postinfarction LV remodeling are uncertain. Accordingly, a well-characterized postinfarction model was used to determine the effects of combined ACEI and AT1 antagonism on postinfarction LV remodeling and LV systolic function.
In 67 female Q fever–negative Dorset sheep (40 to 45 kg), a left thoracotomy was performed with ligation of the left anterior descending coronary artery and its second diagonal branch to create a moderate-size transmural anteroapical infarction as previously described.4 8 11 20 21 Sixty-two sheep completed the protocol (5 sheep died of ventricular arrhythmia on the day of infarction). All procedures followed were in accordance with the “Position of the American Heart Association on Research Animal Use,” adopted by the Association in November 1984, and the Institutional Animal Care and Use Committee at Allegheny General Hospital. Beginning on day 2 after MI, animals were treated with no therapy (Control, n=12), ramipril 10 mg PO QD (standard-dose ACEI; sACEI, n=14), ramipril 10 mg PO BID (high-dose ACEI; hACEI, n=8), losartan 25 mg PO BID (AT1, n=13), or combination therapy with ramipril 10 mg PO QD and losartan 25 mg PO BID (CT, n=15). Previous work in this model has demonstrated that the 10-mg/d dose of ramipril inhibits circulating ACE activity and reliably attenuates LV remodeling.11 The 20-mg/d dose of ramipril was also studied to determine differential effects of high-dose ACEI. The dose of losartan was selected on the basis of its ability to blunt by >50% the pressor response to a 10-minute intravenous infusion of Ang II (at a dose that consistently increased mean arterial pressure by 30% in this animal model) and to maintain this effect against repeated challenges of Ang II up to 11 hours after the last dosing. Serial blood pressures (with an automated pediatric blood pressure cuff in the left forelock of the standing animal) and heart rates were recorded daily for 5 days and then weekly thereafter. Left atrial pressures 8 weeks after MI were also measured.
Magnetic Resonance Imaging
MRI was performed at baseline (preinfarction) and at 8 weeks after infarction. Before the imaging sessions, all animals were premedicated intravenously with diazepam (1 mg), penicillin (22 000 U/kg), and gentamicin (3 mg/kg). Intravenous 5% guaifenesin and 500 mg of ketamine provided heavy sedation during imaging. Intubation, mechanical ventilation, nasogastric suction, and ECG monitoring were performed. The anesthetized, ventilated animal was placed in the right lateral decubitus position on a phased-array surface coil in a Siemens 1.5-T scanner, and ECG gating was initiated.
Multiple–breath-hold, segmented–k-space, multiphase gradient-echo cine series were performed (for measurement of LV mass [LVM], end-systolic and end-diastolic volumes [ESV and EDV], and ejection fraction [EF]) spanning the LV from apex to base. Imaging parameters were repetition time (TR) 60 ms (with view-sharing, yielding a 30-ms temporal resolution), echo time (TE) 4.8 ms, slice thickness 7 mm, 128×256 matrix, and 25-cm field of view, yielding a final interpolated pixel size of 0.95 mm2. Two-chamber and 4-chamber apical long-axis images were also obtained. For analysis of regional intramyocardial function, breath-hold–tagged short-axis images were obtained spanning the LV from apex to base as previously described.4 21 Imaging parameters included 7-mm slice thickness, 7-mm tag stripe separation, TR 70 ms (with view-sharing, yielding a 35-ms temporal resolution), TE 4 ms, 128×56 matrix, 25-cm field of view, final interpolated pixel size 0.95 mm2, and 2 signal averages. The ventilator was held at end expiration during all breath-hold sequences.
MR Image Analysis
LVM, ESV, EDV, and EF were measured from stacked short-axis cine slices by blinded image analysts using Siemens Imageview software (Siemens Corporate Research) and published techniques.4 21 LVM, ESV, and EDV were then indexed to body weight in kilograms (LVMI, ESVI, and EDVI). Infarct size at 8 weeks was measured as the mass of thinned infarcted tissue from interleaved end-diastolic short-axis images and was expressed as a percentage of myocardial mass by previously published techniques that have been demonstrated to correlate highly with a pathological “gold standard” in this model.8 Regional percent intramyocardial segment shortening (%S) from stacked short-axis tagged images was measured by a single blinded observer using a software package (VIDA, © University of Iowa) on a SUN workstation by previously reported methods.4 8 11 21 Interstripe distances were measured at end systole (Les) and end diastole (Led), and %S was calculated as %S=100(Led−Les)/Led. %S was measured at subendocardial and subepicardial sites in segments located within infarcted, adjacent, and remote regions. As defined in previous studies, adjacent regions are those within 2 cm of the clearly demarcated infarct border, and remote regions are >2 cm removed.4 8 11 21 Preinfarct baseline data were analyzed by apex-to-base location and matched to postinfarct images by previously published techniques. Regional myocardial wall thickness (infarct, adjacent, and remote) was measured at 8 weeks after MI from 4-chamber long-axis end-diastolic cine MRI images by blinded observers using previously published techniques.8
Changes between baseline and 8 weeks after MI within each treatment group in blood pressure, heart rate, LVMI, EDVI, ESVI, EF, and regional %S were assessed. Infarct mass, left atrial pressure, and end-diastolic regional myocardial wall thickness were also compared between groups by ANOVA. Data are presented as group mean values±SD, with a value of P<0.05 defined as significant.
Infarct size was similar for all groups (Control, 23±8%; sACEI, 23±7%; hACEI, 19±2%; AT1, 22±5%; and CT, 23±6%; P=NS for all). Baseline, 4-week post-MI, and 8-week post-MI mean blood pressures were similar for Control, sACEI, hACEI, AT1, and CT, respectively: 73±14, 73±15, 72±16, 72±17, and 72±9 mm Hg at baseline; 71±8, 72±9, 70±13, 71±12, and 69±8 mm Hg at 4 weeks; and 71±12, 76±7, 71±13, 69±12, and 70±9 mm Hg at 8 weeks (P=NS for all). Baseline heart rates were similar, but infarcted controls had a higher heart rate than sACEI, hACEI, AT1, or CT at 8 weeks after MI, respectively: 124±17*, 102±12, 104±15, 111±12, and 107±12 bpm (*P<0.0009). Left atrial pressure at 8 weeks after MI was similar for control, sACEI, hACEI, AT1, and CT, respectively: 10±2, 11±2, 10±3, 11±2, and 11±2 mm Hg, P=NS).
Global LV Remodeling and Function
Baseline, 8 weeks after MI, and change from baseline results are shown for the 5 treatment groups (Table 1⇓). At baseline, all parameters were similar between groups. The decline in EF associated with postinfarction LV remodeling was significantly less with CT than with sACEI, AT1 blockade, or control and was also less for hACEI than for control. The increase in EDVI was also less with CT than with AT1 blockade or control (P<0.02), and the increase in ESVI with CT was less than with sACEI, AT1 blockade, or control (P<0.04). In contrast, hACEI changes in EDVI did not differ from other groups, and ΔESVI differed only from control (P<0.05). AT1 blockade showed ΔESVI and ΔEDVI very similar to control, but EF was similar to that with sACEI. Both LV EDVI and ESVI at 8 weeks after MI were significantly lower with CT than with no therapy, sACEI, hACEI, or AT1 blockade. Although the absolute changes in LVMI during the remodeling period were small in all treatment groups, LVMI actually decreased with CT and hACEI. At 8 weeks after MI, LVMI was significantly lower for both CT and hACEI than with no therapy, sACEI, or AT1 blockade.
Regional Wall Thickness
Representative end-diastolic long-axis images are shown in Figure 1⇓ for all treatment groups. End-diastolic regional wall thickness at 8 weeks after MI was similar for control, sACEI, AT1 blockade, CT, and hACEI, respectively: infarct, 2±1, 3±1, 2±1, 3±1, and 3±1 mm; adjacent myocardium, 6±1, 7±1, 6±1, 7±1, and 7±1 mm; and remote myocardium, 9±1, 10±2, 9±1, 10±2, and 10±2 mm (P=NS for all).
Regional LV Function
The results for baseline, 8 weeks after MI, and change from baseline for %S in adjacent noninfarcted myocardial regions are shown in Table 2⇓. At baseline, %S in future adjacent and remote noninfarcted regions was similar for all groups. Normal heterogeneity was noted, with apical and subendocardial regions showing higher %S than basal and subepicardial regions.4 8 11 21 At 8 weeks after MI, adjacent %S in the CT group was higher than all other treatment groups, and the decrease from baseline was significantly less with CT than with sACEI, AT1 blockade, and control. With hACEI, the fall in adjacent %S in subendocardium was less than with sACEI or Control. There were no significant between-group differences in remote myocardial %S at baseline or at 8 weeks after MI.
The present study demonstrates that when AT1 blockade is added to sACEI after transmural anteroapical MI, the combination further attenuates LV remodeling and preserves LV systolic function better than with sACEI alone. The 2 most important prognostic variables predictive of postinfarction mortality, ESVI and EF,22 23 are improved with CT (Figure 2A⇓ and 2B⇓) compared with sACEI alone, AT1 antagonism alone, or control. AT1 antagonism as monotherapy tended to limit postinfarction LV systolic dysfunction, but did not attenuate geometric LV remodeling in this model. Interestingly, treatment with hACEI did provide further benefit compared with sACEI, especially with regard to changes in LVMI. There was a trend favoring CT over hACEI with respect to ΔESVI and ΔEDVI, but this was not statistically significant by ANOVA. Both ESVI and EDVI at 8 weeks after MI were significantly smaller with CT than with all other groups. CT was also superior at limiting adjacent regional systolic dysfunction compared with sACEI, AT1 blockade, or no therapy. Beneficial effects of CT occurred in the absence of any differences in blood pressure or LV filling pressure between treatment groups, suggesting that the differences were related to direct myocardial effects. Myocardial blood flow and flow reserve are unchanged in adjacent and remote regions at 8 weeks after MI21 and therefore would not be expected to play a role in this process. We have previously shown that there is no significant myocardial fibrosis qualitatively in adjacent or remote noninfarcted myocardium in untreated infarcted control animals in this model.8 Quantitative analysis of adjacent and remote region volume percent collagen in untreated infarcted control animals also did not show any evidence of myocardial fibrosis (Sanford P. Bishop, DVM, PhD, unpublished data). Thus, although the myocardial interstitium may play a role in postinfarction remodeling, it is unlikely that differences in the extracellular environment can explain all of the observed results.
Increases in adjacent region LV wall stress in control and sACEI-treated animals, coupled with a reduction in wall stress in the CT group, could contribute to the improved EF and adjacent %S seen with CT. Valid estimation of wall stress in infarcted ventricles with inhomogeneous regional structure and function requires complex finite-element analyses incorporating detailed information on the material properties of each region24 and is very difficult in vivo. Therefore, at present, we could examine only the geometric determinants of wall stress, namely, wall thickness and ventricular volume. Nonetheless, our previous studies have hypothesized that elevated wall stress is an important determinant of chronic adjacent-regional dysfunction.8 Other indirect evidence of increased regional wall stress is derived from studies assessing wall curvature on a regional basis in the chronic postinfarct LV.25 26 Regional thickness and blood pressure were similar between treatment groups in this study, but EDV and ESV (and hence ventricular radii) are reduced with CT compared with controls, AT1 blockade, and sACEI. Thus, wall stress must be lower by the Laplace relationship and could contribute to improved EF and adjacent %S.
Monotherapy with ACEI has consistently been shown to limit postinfarction LV remodeling in experimental studies as well as in clinical trials. Both direct and indirect actions of the renin-angiotensin system have been implicated in this process. Depending on the model studied, conflicting data exist on the relative importance of reductions in Ang II levels and inhibition of bradykinin breakdown in this process.14 15 27 28 The ability of ACEI to decrease Ang II synthesis may also be diminished over time by the production of Ang II via non-ACE pathways within the myocardium.16 29 In fact, the compensatory increase in Ang I associated with ACEI may drive non-ACE production of Ang II, as evidenced by the nearly normal levels of Ang II found in heart failure patients treated chronically with ACEI.30 Although the present study suggests that AT1 antagonism provides additional benefits when added to ACEI after myocardial infarction, it remains unclear whether AT1 antagonism has an independent effect or merely potentiates the effects of ACEI by offsetting a variety of compensatory mechanisms that normally blunt the effectiveness of ACEI alone.
Comparison With Previous Studies
Conflicting data exist on the effect of AT1 blockade alone on postinfarction LV remodeling. In a rat coronary ligation model, Schieffer et al14 found that ACEI and AT1 blockade were equally effective in limiting postinfarction LV remodeling. Milavetz et al28 confirmed these findings and showed no difference in survival at 1 year between rats treated with ACEI or an AT1 antagonist after coronary ligation–induced MI. In an electrical injury canine model, however, McDonald et al27 found that AT1 blockade failed to attenuate LV remodeling.
To the best of our knowledge, the effect of combined ACEI and AT1 blockade on postinfarction LV remodeling has not previously been evaluated. Combination therapy, however, has been evaluated in other experimental and clinical conditions. In a porcine model of rapid pacing–induced congestive heart failure, Spinale et al17 found that combination therapy provided greater preservation of LV pump function and geometry than ACEI alone. LV end-diastolic dimension was smaller and LV fractional shortening, velocity of circumferential fiber shortening, and cardiac output were all greater with combination therapy than with ACEI alone. These results are consistent with the findings of the present study. Interestingly, AT1 antagonism alone did not have a beneficial effect on genetic remodeling in this model. Recently, the addition of AT1 blockade to recommended or maximally tolerated ACEI in patients with heart failure was shown to improve exercise capacity and reduce symptoms.19 In Val-HeFT,31 however, the addition of AT1 blockade to standard heart failure therapy did not improve mortality despite significantly improving EF, quality of life, and NYHA functional class.
The 10-mg/d dose of ramipril used in the sACEI group was based on previous work in this animal model demonstrating consistent inhibition of ACE activity and attenuation of postinfarction LV remodeling.11 The 20-mg/d dose of ramipril used in the hACEI group was arbitrary, however, and further effects with even higher doses of ramipril cannot be excluded. The present study does not provide a complete description of the dose-response curves for ACEI, AT1 antagonism, and their combination in this animal model. Further studies to more completely describe such dose-response curves would provide important information. The volumetric accuracy and superior reproducibility of cardiac MRI have consistently provided adequate statistical power with similar sample sizes in this model4 8 11 21 and other models.15 17 Nevertheless, the number of study groups did limit statistical power in the present study. Only 1 time point during remodeling was studied. The 8-week-postinfarction time point was chosen on the basis of the time course of remodeling previously demonstrated in this model.4 8 11 21 Effects beyond this time point may warrant further investigation. Finally, the cellular mechanisms responsible for the additive effects seen with combined ACEI and AT1 blockade in the present study remain unclear.
A large, multicenter trial that will compare ACEI plus AT1 blockade, ACEI alone, and AT1 blockade alone in humans after MI is currently under way. One of the hypotheses underlying this trial is that combined therapy is superior to standard doses of ACEI alone. The results of the present study support that hypothesis.
This study was supported by HL-52980 (Dr Kramer), Grant-in-Aid 96014830 from the American Heart Association National Office (Dr Kramer), and a Losartan Medical School Grant from Merck Inc (Dr Mankad and Dr Kramer). The authors gratefully acknowledge Therese M. Theobald, MPH, and Amy L. Shaffer, MS, for their technical support, and Hoechst Marion Roussel, Inc for supplying ramipril.
- Received October 5, 2000.
- Revision received February 9, 2001.
- Accepted February 15, 2001.
- Copyright © 2001 by American Heart Association
Eaton LN, Bulkey BH. Expansion of acute myocardial infarction: its relationship to infarct morphology in a canine model. Circ Res. 1981;49:80–88.
Pfeffer MA, Braunwald E. Ventricular remodeling after myocardial infarction: experimental observations and clinical implications. Circulation. 1990;81:1161–1172.
Anversa P, Beghi C, Kikkawa Y, et al. Myocardial infarction in rats: infarct size, myocyte hypertrophy, and capillary growth. Circ Res. 1986;58:26–37.
Olivetti G, Capasso JM, Sonnenblick EH, et al. Side-to-side slippage of myocytes participates in ventricular wall remodeling acutely after myocardial infarction in rats. Circ Res. 1990;67:23–34.
Kramer CM, Lima JAC, Reichek N, et al. Regional differences in function within noninfarcted myocardium during left ventricular remodeling. Circulation. 1993;88:1279–1288.
Pfeffer JM, Pfeffer MA, Braunwald E. Influence of chronic captopril therapy on infarcted left ventricle of the rat. Circ Res. 1985;57:84–95.
St John Sutton M, Pfeffer MA, Plappert T, et al. Quantitative two-dimensional echocardiographic measurements are major predictors of adverse cardiovascular events after acute myocardial infarction. Circulation. 1994;89:68–75.
Schieffer B, Wirger A, Meybrunn M, et al. Comparative effects of chronic angiotensin-converting enzyme inhibition and angiotensin II type 1 receptor blockade on cardiac remodeling after myocardial infarction in the rat. Circulation. 1994;89:2273–2282.
McDonald KM, Mock J, D’Aloia A, et al. Bradykinin antagonism inhibits the antigrowth effect of converting enzyme inhibition in the dog myocardium after discrete transmural myocardial necrosis. Circulation. 1995;91:2043–2048.
Hollenberg NK, Fisher NDL, Price DA. Pathways for angiotensin II generation in intact human tissue. Hypertension. 1998;32:387–392.
Urata H, Healy B, Stewart RW, et al. Angiotensin II–forming pathways in normal and failing human hearts. Circ Res. 1990;66:883–890.
Spinale FG, de Gasparo M, Whitebread S, et al. Modulation of the renin-angiotensin pathway through enzyme inhibition and specific receptor blockade in pacing-induced heart failure: effects on left ventricular performance and neurohormonal systems. Circulation. 1997;96:2385–2396.
Hamroff G, Katz SD, Mancini D, et al. Addition of angiotensin II receptor blockade to maximal angiotensin-converting enzyme inhibition improves exercise capacity in patients with severe congestive heart failure. Circulation. 1999;99:990–992.
Jorde UP, Ennezat PV, Lisker J, et al. Maximally recommended doses of angiotensin-converting enzyme (ACE) inhibitors do not completely prevent ACE-mediated formation of angiotensin II in chronic heart failure. Circulation. 2000;101:844–846.
White HD, Norris RM, Brown MA, et al. Left ventricular end-systolic volume is the major determinant of survival after recovery from myocardial infarction. Circulation. 1987;76:44–51.
Multicenter Post-Infarction Research Group. Risk stratification and survival after myocardial infarction. N Engl J Med. 1983;309:321–336.
Yin FCP. Ventricular wall stress. Circ Res. 1981;49:829–842.
Janz RF, Waldron RJ. Predicted effect of chronic apical aneurysms on the passive stiffness of the human left ventricle. Circ Res. 1978;42:255–263.
Janz RF. Estimation of local myocardial stress. Am J Physiol. 1982;242:H875–H881.
McDonald KM, Garr M, Carlyle PF, et al. Relative effects of α1-adrenoreceptor blockade, converting enzyme inhibitor therapy, and angiotensin II subtype 1 receptor blockade on ventricular remodeling in the dog. Circulation. 1994;90:3034–3046.
Wolny A, Clozel J-P, Rein J, et al. Functional and biochemical analysis of angiotensin II–forming pathways in the human heart. Circ Res. 1997;80:219–227.
Wei CC, Meng QC, Palmer R, et al. Evidence for angiotensin-converting enzyme– and chymase-mediated angiotensin II formation in the interstitial fluid space of the dog heart in vivo. Circulation. 1999;99:2583–2589.
Cohn JN. Val-HEFT (the Valsartan Heart Failure Trial). Clin Cardiol. 2001;24:86.