Effects of Long-term Enalapril Therapy on Cardiac Structure and Function in Patients With Left Ventricular Dysfunction
Results of the SOLVD Echocardiography Substudy
Background Studies of Left Ventricular Dysfunction (SOLVD) demonstrated that enalapril therapy significantly improved the clinical course of patients with left ventricular (LV) dysfunction. The goals of this substudy were to evaluate changes in LV structure and function in SOLVD patients and to test the hypothesis that enalapril inhibits remodeling in patients with LV dysfunction.
Methods and Results Patients entering both the prevention and treatment arms of SOLVD from 5 of the 23 clinical centers were recruited for this substudy. The 301 patients who participated underwent Doppler-echocardiographic evaluation according to standard protocol before randomization to either enalapril or placebo and again after 4 and 12 months of therapy. Recorded data were analyzed in a blinded fashion at a central core laboratory. Analysis of baseline clinical characteristics showed that patients enrolled in the substudy were generally representative of the SOLVD population, although prevention arm patients were slightly overrepresented in the substudy group (69.8% compared with 61.9% of remaining SOLVD patients). The enalapril group demonstrated significant reductions in the mitral annular E-wave–to–A-wave velocity ratio (due predominantly to a reduction in E-wave velocity), and this response was different from that seen in the placebo group (P=.030). Changes in the E-to-A ratio in the enalapril group correlated significantly with changes in plasma atrial natriuretic peptide (r=.56; P≤.01). LV end-diastolic and end-systolic volumes increased in placebo but not enalapril-treated patients, and the differences in response between the treatment groups were significant (P=.025 and .019, respectively). LV mass tended to increase in placebo patients and to be reduced in enalapril-treated patients, and the difference in response between the groups was highly significant (P≤.001).
Conclusions These data demonstrate that enalapril attenuates progressive increases in LV dilatation and hypertrophy in patients with LV dysfunction. The results support the possibility that the favorable effects of enalapril reported in the SOLVD trials were related to inhibition of LV remodeling.
Patients with congestive heart failure (CHF) due to left ventricular (LV) systolic dysfunction experience a limitation in activity due to symptoms and a substantial reduction in survival rate.1 Evidence that activation of the renin-angiotensin system is involved in the pathogenesis of CHF resulted in the testing of angiotensin-converting enzyme inhibitor (ACE-I) agents in this setting.2 The results of Studies of Left Ventricular Dysfunction (SOLVD) and other long-term, placebo-controlled trials demonstrate that ACE-I therapy significantly improves the clinical course of a broad spectrum of patients with LV dysfunction.3 4 5 6 7
Increases in LV chamber volumes and muscle mass are an important consequence of LV dysfunction. Although changes in myocardial architecture help maintain cardiac performance in the face of altered hemodynamic loading conditions and diminished pump capacity,8 9 the remodeling process is associated with the late onset of progressive LV dysfunction and decreased survival.9 10 11 12 13 Previous studies have suggested that when ACE-I therapy is begun shortly after an acute myocardial infarction, progressive LV dilatation can be inhibited.14 15 However, the effects of ACE-I therapy on LV structure and function in patients with more long-standing LV dysfunction is uncertain. As part of SOLVD, combined Doppler-echocardiographic evaluation was performed longitudinally in a subset of patients recruited from both the prevention and treatment arms of the trial. The goals of this substudy were to evaluate the long-term effects of enalapril therapy on changes in the LV in patients with chronic systolic dysfunction. In particular, the investigators sought to determine whether ventricular remodeling occurred in SOLVD patients and whether this process was altered by treatment with enalapril.
Five of the 23 SOLVD clinical centers participated in the echocardiographic substudy. These centers were located at the Oregon Health Sciences University, Portland, Ore; Baylor College of Medicine, Houston, Tex; Dalhousie University, Halifax, Nova Scotia, Canada; University of Florida College of Medicine (Gainesville); and the Robert Wood Johnson School of Medicine, Piscataway, NJ. Tapes of the Doppler-echocardiographic studies were sent to a core laboratory at the Baylor College of Medicine, where the readers had no access to clinical data, previous results (in the case of follow-up studies), or the treatment to which the patient had been randomized. The results were entered into a computerized database at the Central Coordinating Center located at the University of North Carolina at Chapel Hill. The study protocols were approved by the institutional review boards at each of these centers, and informed consent was obtained from all patients.
A detailed description of the design of SOLVD, including entry criteria and randomization strategy, has been published.16 Briefly, patients between 21 and 80 years of age were eligible for the study if they had a LV ejection fraction measurement ≤0.35 within the preceding 3 months. Patients could not be enrolled into SOLVD within 30 days of an acute myocardial infarction. Patients fulfilling entry criteria were initially administered 2.5 mg enalapril BID in an unblinded fashion for 2 to 7 days followed by administration of placebo for 14 to 17 days. At the time of randomization, patients were assigned to either the treatment or prevention arms of SOLVD depending on whether signs and symptoms of overt CHF requiring treatment were either present or absent. In both arms of SOLVD, enalapril or placebo therapy was then initiated in a double-blinded fashion beginning (in most patients) at a dose of 5 mg BID and, as tolerated, increased to 10 mg BID. Patients were followed at regular intervals for an average of 39.2 and 37.4 months, respectively, in the treatment and prevention arms of SOLVD.
All patients enrolled in the main SOLVD study at the five substudy clinical centers during the period of late 1987 through 1990 were considered for inclusion in the echocardiographic substudy if they were willing to participate. The calculated target sample size based on the expected changes in LV end-diastolic volume over 12 months in the treatment groups and the variability in echocardiographic measurements was 300 patients. Baseline Doppler-echocardiographic studies were obtained within 24 hours of the eligibility visit and before the initial single-blind challenge with 2.5 mg enalapril. Patients whose echocardiograms were considered by predefined criteria to be of insufficient quality for quantitative analysis were excluded from further participation in the substudy. Follow-up evaluation was performed at SOLVD visits 4 and 12 months after study medication was begun.
Efforts to reduce variability between studies included the use of a standardized protocol, recording of the patient’s position and interspace(s) used at the initial study for future reference, and performance of the baseline and follow-up Doppler-echocardiographic studies at the same time of day in individual patients. Patients were instructed to take their study medication on the day of the evaluation and to consume only clear liquids in the 4-hour period before the Doppler-echocardiographic study.
An identical Doppler-echocardiographic examination was performed on a control population consisting of healthy subjects at each of the participating centers. These individuals were without any history of cardiovascular disease, and none were receiving medications for such a condition. The 53 patients who comprised the control group averaged 52±15 years of age. There were 28 (54%) men and 25 (45%) women. Tapes of these studies were sent to the core laboratory at the Baylor College of Medicine for analysis.
Echocardiographic studies were performed using commercially available equipment with two-dimensional and pulsed Doppler capabilities. Before the start of the trials, the sonographers from the five participating centers underwent a training session to standardize the imaging planes and recording techniques. All centers completed standardized phantom imaging with good reproducibility. Patients were studied in the left lateral recumbent position. Systemic blood pressure was measured with a cuff with the patient in the supine position before examination after a 2-minute rest.
The imaging planes consisted of the parasternal long- and short-axis views followed by the apical four- and two-chamber views. From the apical window, a pulsed Doppler cursor was positioned in the mitral valve inflow region with the sample volume placed at the level of the mitral annulus to detect mitral regurgitation and record the annular inflow velocities. All Doppler velocities were recorded at a sweep of 100 mm/s.
Echocardiographic studies were recorded on videotape and forwarded to the SOLVD echocardiographic core laboratory. Representatives cardiac cycles were captured on digital format at 50-millisecond intervals and stored on floppy disks. Measurements were performed on a computer analysis station that had been calibrated. All measurements were made by experienced observers who were blinded to clinical data and previous studies. Whenever possible, results were expressed at an average of three cardiac cycles in sinus rhythm and three cycles representatives of the average RR interval in atrial fibrillation.
The dimension of the left atrium was measured at end systole from an M-mode recording at the level of the aortic root.17 Measurements of cardiac chambers were taken from the two-dimensional images using inner edge to inner edge. End diastole was defined as the frame coinciding with the onset of the QRS on the ECG, and end systole was defined as one frame before the opening of the mitral valve. Measurements included LV end-diastolic and end-systolic diameters at the upper and mid thirds of the LV cavity in the parasternal long-axis view and at the upper, mid, and lower thirds of the LV cavity in the apical four- and two-chamber views. Septal and posterior wall thicknesses were measured at end diastole at the level of the upper diameter in the parasternal view. The outer and inner contours of the LV (excluding the papillary muscles) were traced at end diastole and end systole from the short-axis view at midpapillary muscle level. The LV long axis was measured in each apical view from the outermost border of the apical endocardium to the center of the mitral annulus plane. The longer of the two measurements was used as representative of the long axis. The left and right atrial cavities were traced at end systole in the four-chamber view to derive an area estimate of atrial sizes.
All Doppler recordings were digitized along the outer border of the brightest portion of the velocity contour. Peak early (E) and atrial (A) velocities were measured, and atrial filling fraction was derived as the ratio of the integral of the A wave to the total integral of the velocity. Mitral inflow velocities were not analyzed in the presence of a tachycardia that induced merging of E and A velocities. In atrial fibrillation, only the E velocity and deceleration time were measured. The isovolumic relaxation time was measured as the interval from the end of LV outflow velocity to the onset of mitral inflow velocity.
LV volumes were derived using the multiple-diameter method.18 LV mass was calculated by the cube formula19 using the diastolic cavity diameter and the septal and posterior wall thicknesses from the parasternal view. This method was selected to increase the yield of studies in which mass could be determined since it does not depend on the quality of the short-axis or apical views. Comparison of the cube formula with the area–length method in 36 patients with studies of excellent quality revealed a direct linear correlation with a correlation coefficient of .94 (P<.001) and 95% confidence limits of 80 g. The equation describing this relation is y=0.67x+29.7 where y represents the area length method and x represents the cube formula method.
Systolic meridional wall stress and circumferential wall stress were derived by combining systolic blood pressure with echocardiographic measurements at end systole obtained from the parasternal and apical views using the modified equations described by St John Sutton et al.20
The reproducibility of individual measurements and derived parameters at the core laboratory, expressed as the mean±1 SD difference between two observers, is 0.05±0.10 cm for dimensions; 1±14 mL and 9±25 g for LV volumes and mass, respectively; 0.11±0.16 for E-to-A ratio; and 2.9±2% and 8.1±7.6 milliseconds for atrial filling fraction and isovolumic relaxation time, respectively.
An attempt was made to obtain measurements from all of the echocardiographic studies. However, at times the quality of the images was suboptimal for a certain measurement because of either the tangential tomographic planes or loss of endocardial resolution. Consequently, not all measurements were available in every patient. The mitral inflow velocities and atrial filling fractions were determined in 83% of patients; in 11% and 6%, the analysis was precluded by sinus tachycardia or atrial fibrillation, respectively. Left atrial and right atrial areas could be measured in 61% and 50%. Measurements of chamber dimensions, LV volumes, and LV mass were obtained in 92%, 85%, and 85% of the patients, respectively. Estimates of systolic wall stress were possible in 67% of the patients.
Centers participating in the echocardiographic substudy were encouraged to enroll patients in the neurohormonal substudy so that both sets of measurements could be obtained in a subset of patients. Selection of patients was random and depended primarily on the logistics at each clinic. Blood was obtained at the time of the baseline and 12-month follow-up studies in a standardized manner according to methods that have been described previously.21 Measurements of plasma norepinephrine (PNE), plasma renin activity (PRA), atrial natriuretic peptide (ANP), and arginine vasopressin (AVP) were performed by techniques that are described in detail elsewhere.22 23 24 All samples were analyzed in a blinded fashion without knowledge of the patient’s treatment, study drug assignment, or clinical status. A smaller number of patients had blood drawn for ANP and AVP determinations than for the other neurohormones by design of the neurohormonal study group.21
All statistical analyses were performed at the SOLVD coordinating center. Differences in baseline prerandomization characteristics between echocardiographic substudy and remaining SOLVD patients and between substudy patients randomized to enalapril and placebo were evaluated using either Fisher’s exact test, Pearson’s χ2 analysis, or the Wilcoxon two-sample test.
The effects of treatment on heart rate, blood pressure, and Doppler-echocardiography variables were analyzed according to the intention-to-treat principle so that all patients who underwent baseline and at least one follow-up evaluation were included regardless of whether they were receiving the study drug at the time of the 4- and 12-month studies. Statistical analysis included comparison of the changes within the placebo and enalapril groups at 4 and 12 months and a comparison of the response between treatment groups over time. Evaluation of the association between changes in the variables over time was restricted to the analyses included in the report to minimize the possibility of type II error (ie, falsely concluding that a significant relationship exists between variables).
The changes in the treatment groups over time were compared by repeated-measures analysis. Because this analysis requires that data from all three longitudinal measurements be present, missing data points (due to death, absence from study visit, or poor-quality data) were generated using an imputation algorithm in cases in which data from two of the three study visits was available.25 Patients in whom only baseline and 4-month measurements were available were assigned an imputed value for the 12-month visit that was equal to the 4-month value. Patients with baseline and 12-month measurements received a 4-month imputed value calculated from the slope of the line relating the baseline and 12-month visits. The percentages of imputed values from both study visits for mitral flow velocities, LV volumes, and LV mass were 36%, 31%, and 37%, respectively. Overall, analyses using the imputation algorithm were similar to repeated-measures analysis in which only patients with all three data points were available. However, because the latter would exclude considerable amounts of useful data, the imputed values are included in this report. The nonimputed values are included as the Table⇓ in the “Appendix.”
To determine whether assignment to the treatment or prevention arm of SOLVD affected the response to therapy, it was decided to include treatment by trial (ie, treatment or prevention) interactions in the initial repeated-measures model. Treatment by baseline covariate interactions were also included, making this a repeated-measures covariance analysis in which the 4- and 12-month values were modeled as responses and the baseline variable was used a covariate.26 To account for baseline inhomogeneities (Table 2⇓) and the fact that data were collected at five separate study centers, age, sex, and clinic site were also included as covariates in the model. When a treatment-by-trial interaction was detected, the nature of the interaction was examined for the 4- and 12-month studies separately. Only systolic and diastolic blood pressures were affected at different follow-up times by the treatment-by-trial interactions, and results for each trial are reported separately for these variables.
Descriptive variables from the baseline evaluation of the 301 patients randomized in the echocardiographic substudy are compared with those obtained in the remaining SOLVD patients (Table 1⇑). The proportion of prevention-arm patients enrolled into the substudy was slightly greater than in the remaining SOLVD patients (69.8% versus 61.9%), whereas the percentage of treatment-arm patients was lower (30.2% versus 38.1%). As a result, substudy patients were more likely than the remaining SOLVD patients to be New York Heart Association functional class I. A slightly higher percentage of substudy patients than remaining SOLVD patients were receiving β-blockers, calcium channel blockers, and vasodilators at the time of entry. Substudy patients also had slightly higher ejection fractions (determined at each center before entry into the trial) and heart rates than did the remaining SOLVD patients. The actual differences in all these variables, however, was small. Other characteristics, including sex, age, etiology of LV dysfunction, use of digitalis and diuretics, and blood pressure, were almost identical in the groups.
As indicated in Table 2⇑, the number of patients in the echocardiography substudy who were randomized to enalapril was similar to the number randomized to placebo. Comparison of baseline characteristics in substudy patients according to the treatment assigned revealed that patients in the placebo group were more likely to be male (90.0% versus 80.4%; P=.035) and were slightly older (60.4±0.8 versus 58.0±0.9 years; P=.074) than patients in the enalapril group. All other baseline clinical characteristics were similar in the two treatment groups.
Effects of Therapy on Arterial Pressures and Heart Rate
A summary of the changes in blood pressure in echocardiographic substudy patients and comparisons of these changes between the treatment groups are given in Table 3⇓. The results for systolic and diastolic pressures are reported separately for patients in the two arms of SOLVD since the response in prevention-arm patients differed from that in treatment-arm patients at different time points. In general, enalapril had a more significant effect in reducing blood pressure in prevention-arm patients at 12 months, whereas its effect was more significant in the treatment-arm patients at the time of the 4-month follow-up visit. As shown in Table 4⇓, heart rate did not change significantly in either treatment group.
The variables measured from the Doppler-echocardiographic studies in the SOLVD patients and in the healthy control subjects are summarized in Table 4⇑. Of note are the substantial increases in left atrial dimension and systolic area, isovolumic relaxation time, end-diastolic and end-systolic volumes, LV mass, and meridional and circumferential wall stress in the SOLVD patients. At baseline, the enalapril group differed from the placebo group only in the measurement of isovolumic relaxation time (P=.033) and LV mass (P=.021).
Variables Related to LV Filling
The effects of therapy on the Doppler-echocardiographic variables associated with LV filling are summarized in Table 4⇑. Over the course of the study, patients in the enalapril group demonstrated highly significant reductions in mitral annular E-wave velocity and small increases in A-wave velocity at both 4 and 12 months compared with baseline. Since this variable also tended to be reduced in the placebo patients, however, comparison between the groups over time was only of borderline significance for E-wave velocity and was not significant for A-wave velocity. Mitral annular E-to-A velocity ratio was significantly reduced in the enalapril but not in the placebo group at both 4 and 12 months, and the difference in response between these groups was significant (P=.03). Atrial filling fraction increased in the enalapril-treated patients but not in the placebo-treated patients. Isovolumic relaxation time fell at 4 and 12 months in the placebo group, but the changes over time were not significantly different from those measured in the enalapril group.
An interaction between the response to therapy and arm of SOLVD was detected for changes in mitral annular E-wave velocity, E-wave–to–A-wave ratio, and filling fraction. For these variables, changes in the enalapril group were greater in the patients followed in the treatment arm of SOLVD than in the patients followed in the prevention arm. For all other variables, the response was similar in the two arms of SOLVD.
Left atrial dimension was slightly but significantly reduced in the enalapril-treated but not in the placebo-treated patients at the 4-month visit. Only small and insignificant changes in left atrial and right atrial systolic areas were noted in the study groups.
Relation Between Mitral E-to-A Ratio and Neurohormonal Measurements
Spearman correlation coefficients between the change in mitral E-to-A ratio from the baseline to the 12-month study and the changes in PNE (n=99), PRA (n=98), and AVP (n=43) were −.06, .06, and −.03, respectively (all P=NS). The correlation between changes in mitral E-to-A ratio and ANP levels (n=43) was .40 (P=.008). This correlation appeared to be due almost entirely to the association in the enalapril-treated patients (r=.56; P=.01). There was an insignificant correlation in placebo-treated patients (r=.21; P=.34). There was no evidence of a significant association between changes in E-to-A ratio and change in either PNE, PRA, or AVP for either enalapril- or placebo-treated patients.
LV Size, Function, and Wall Stress
As shown in Table 4⇑ and the Figure⇓, placebo-treated patients experienced increases in both LV end-diastolic volume and end-systolic volume over the 1-year period of observation. In both cases, the changes were significant in comparison to changes measured in the enalapril-treated patients. In the latter group, neither variable increased significantly from the baseline level over the 1-year period of treatment. LV ejection fraction remained essentially unchanged in both treatment groups.
Changes in LV mass are also summarized in Table 4⇑ and the Figure⇑. There was a tendency for this variable to increase in the placebo group from 280±100 g at baseline to 297±100 g at the 1-year visit. In comparison, the enalapril-treated group demonstrated a small reduction from 265±82 to 255±82 g over this period. The difference in response between the two treatment groups was highly significant (P<.001).
LV meridional wall stress was reduced in patients in the enalapril-treated group, and there was also a somewhat smaller but still significant reduction in the placebo group. Consequently, differences in responses between the study groups over time were not significant. Circumferential wall stress, however, tended to increase in the placebo group and to be reduced in the enalapril group so that the comparison of the response between the groups demonstrated a significant difference (P=.014). There were no significant interactions between the arm of SOLVD to which patients had been assigned and the changes with treatment for either LV volumes, mass, or wall stress.
Results from SOLVD (as well as those from other large-scale clinical trials) have shown that ACE-I therapy can substantially alter the clinical course of patients with LV dysfunction.3 4 5 6 7 The major goal of the echocardiographic substudy was to assess the effects of enalapril on LV structure and function in patients randomized in SOLVD. The most important finding of the present report is that enalapril attenuated the progressive LV dilatation and hypertrophy that occurred in the placebo group. Since increases in LV volumes and muscle mass are related to future deterioration in LV performance and a less favorable clinical course,9 10 11 12 13 it is likely that the effects of enalapril on the remodeling process are related to the clinical benefits observed in the SOLVD trials.
The patients included in the echocardiographic substudy are generally representative of the patient population randomized in the SOLVD trials. As in SOLVD, the population consisted of mostly middle-aged, white men in whom the etiology of LV dysfunction was coronary artery disease. The slight preponderance of prevention-arm patients who were enrolled in the substudy resulted in differences in New York Heart Association functional class distribution, pattern of drug use, heart rate, and ejection fraction between substudy and remaining SOLVD patients. These differences were quite small, however, and it seems reasonable, based on the overall similarities in baseline characteristics, to relate the results of the substudy to those of the main trial.
The beneficial effects of enalapril on LV dilatation seen in this subgroup of SOLVD patients are similar to those reported in smaller groups of asymptomatic myocardial infarction survivors who were treated with captopril14 15 and symptomatic SOLVD patients in whom LV volumes were assessed with radioisotope or contrast angiography.27 28 In the present study, the effects of enalapril on LV volumes were independent of the arm of SOLVD in which the patient was enrolled, suggesting that therapy was effective regardless of symptomatic status. The large sample size, inclusion of both asymptomatic and symptomatic patients with depressed ejection fractions, and exclusion of patients until at least 1 month after myocardial infarction extends the conclusions of previous studies14 15 and demonstrates that ACE-I therapy attenuates LV dilatation in a broad range of patients with LV dysfunction. In addition, these results are unique in demonstrating that changes in LV mass over time were significantly altered by enalapril therapy. It is noteworthy that in this study enalapril therapy was begun at a time when considerable LV dilatation and hypertrophy (in comparison to a control population) was already present, demonstrating both the chronic progressive nature of the remodeling process and the opportunity for effective intervention even after considerable structural changes have developed.
Although remodeling helps the LV adapt to alterations in loading conditions and serves to maintain cardiac performance in the face of systolic dysfunction,8 9 adverse consequences of this process have been described.9 10 11 12 13 29 30 The strong association between LV mass and outcome was recently emphasized in a preliminary report of 1172 patients from the SOLVD Trials and Registry31 that showed that patients whose LV mass was 1.5 SD above the mean for the group experienced excesses in mortality and cardiovascular hospitalizations of 37% and 28%, respectively. Both increases were highly significant (P<.003) and independent of the effects of ejection fraction. Therefore, interventions that inhibit the remodeling process would be expected to have favorable effects on the natural history of patients with LV dysfunction.
Although the mechanisms by which enalapril inhibits progressive increases in LV end-diastolic and end-systolic volumes and LV mass in patients with LV dysfunction have not been fully resolved, the results of the echocardiographic substudy provide some plausible explanations. LV systolic meridional wall stress was reduced by enalapril, and changes in circumferential wall stress were significantly different from those seen in the placebo group. Since systolic wall stress is believed to be a potent stimulus for the development of LV dilatation and hypertrophy,32 these results suggest that a reduction in wall stress inhibited the progressive LV remodeling that was seen in the patients treated with placebo. However, the results describing wall stress should be interpreted with caution since the presence of segmental wall motion abnormalities might limit the validity of the measurements in a population in which LV dysfunction is due to coronary artery disease, as was largely the case in patients included in the present study. There also is information that angiotensin II stimulates growth of cardiac myocytes33 34 and production of collagen by fibroblasts. Thus, a reduction in the levels of angiotensin II in the heart with enalapril therapy could have contributed directly to the inhibition of cardiac remodeling.
Over the course of the study, a significant reduction in mitral annular E-to-A ratio was observed in patients treated with enalapril. This index of LV filling rate is affected by multiple factors, including the transmitral pressure gradient, heart rate, and the diastolic properties of the LV. The fact that the reduction in E-to-A ratio was predominantly caused by a reduction in mitral E-wave velocity with relatively little change in A-wave velocity and that heart rate was shown to have remained fairly constant during this period suggests that the changes in E-to-A ratio were due to a reduction in left atrial pressures in patients treated with enalapril.35 36 37 Since ANP is predominantly expressed and released from the atria in response to mechanical factors such as increased pressure,38 39 40 the significant correlation between the changes in mitral E-to-A ratio and plasma levels of ANP in the enalapril-treated patients further supports the notion that left atrial pressures were reduced in the enalapril group. The decrease in left atrial dimension in these patients is also consistent with this explanation. This effect of enalapril on LV filling pressures is likely to have resulted in a reduction in diastolic wall stress and could have contributed to the inhibition of LV dilatation and hypertrophy.
In contrast to the changes that were observed in LV structure, no differences were seen between the enalapril- and placebo-treated groups in LV ejection fraction. Since this measurement of systolic performance is load dependent, it may be that a reduction in wall stress, which would favor an increase in ejection fraction, was offset by a reduction in LV filling pressure, which might reduce this variable. In addition, small changes in ejection fraction may be difficult to detect by echocardiography. Assessment of cardiac performance in similar groups of SOLVD patients by radioisotope and contrast angiography did detect small but significant changes in LV ejection fraction in favor of the enalapril-treated group.27 28
Echocardiographic measurements are subject to errors induced by poor image quality or foreshortening of the LV cavity by the tomographic plane, particularly in the apical views. The likelihood of these errors occurring was reduced, we hope, by the training session conducted at the initiation of the study. The formulas used to derive LV volumes, mass, and systolic wall stress indexes are subject to errors in the presence of regional wall motion abnormalities. These abnormalities, however, should be distributed equally between the treatment groups and would not be expected to alter the significance of the changes that were observed in response to therapy. Finally, since changes in Doppler-echocardiographic variables over the period of observation may not be linear, use of an algorithm in which missing values at the 4-month visit are imputed by linear interpolation between the baseline and 12-month values may actually underestimate changes that occurred in the study groups. Nevertheless, comparison of the results in Table 4⇑, which uses imputed values, with those from the Appendix, which uses only nonimputed values, demonstrates that this effect is minimal.
Overall, the results of the present study demonstrate that ACE-I therapy with enalapril inhibits the progression of LV dilatation and hypertrophy in patients with LV dysfunction. Since the study patients are representative of those included in SOLVD, these results provide important insights into the mechanism through which enalapril improves the clinical course of patients with LV dysfunction.
This work was supported by National Heart, Lung, and Blood Institute contract N01-HC-55011, National Institutes of Health, Bethesda, Md.
Guest editor was Myron L. Weisfeldt, MD, Columbia Presbyterian Medical Center, New York, NY.
- Received September 14, 1994.
- Revision received November 28, 1994.
- Accepted December 3, 1994.
- Copyright © 1995 by American Heart Association
Ho KKL, Anderson KM, Kannel WB, Grossman W, Levy D. Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation. 1993;88:107-115.
Curtiss L, Cohn JN, Vrobel T, Franciosa JA. Role of the renin-angiotensin system in the systemic vasoconstriction of chronic heart failure. Circulation. 1978;58:763-771.
Pfeffer MA, Braunwald E, Moye LA, Basta L, Brown EJ, Cuddy TE, Davis BR, Geltman EM, Goldman S, Flaker GC, Klein M, Lamas GA, Packer M, Rouleau J, Rouleau JL, Rutherford J, Wertheimer JH, Hawkins M, on behalf of the SAVE Investigators. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med. 1992;327:669-677.
Ginzton LE, Conant R, Rodrigues DM, Laks MM. Functional significance of hypertrophy of the noninfarcted myocardium after myocardial infarction in humans. Circulation. 1989;80:816-822.
Hammermeister KE, DeRouen TA, Dodge HT. Variables predictive of survival in patients with coronary disease: selection by univariate and multivariate analyses from the clinical, electrocardiographic, exercise, arteriographic, and quantitative angiographic evaluations. Circulation. 1979;59:421-430.
White HD, Norris RM, Brown MA, Brandt PWT, Whitlock RML, Wild CJ. Left ventricular end-systolic volume as the major determinant of survival after recovery from myocardial infarction. Circulation. 1987;76:44-51.
Tortoledo FA, Quinones MA, Fernandez GC, Waggoner AD, Winters WL Jr. Quantification of left ventricular volumes by two-dimensional echocardiography: a simplified and accurate approach. Circulation. 1983;67:579-584.
Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation. 1977;55:613-618.
Francis GS, Benedict C, Johnstone DE, Kirlin PC, Nicklas J, Liang C, Kubo SH, Rudin-Toretsky E, Yusuf S, for the SOLVD Investigators. Comparison of neuroendocrine activation in patients with left ventricular dysfunction with and without congestive heart failure. Circulation. 1990;82:1724-1729.
Hussain MN, Benedict CR. Radioenzymatic microassay for simultaneous estimations of dopamine, norepinephrine and epinephrine in plasma, urine and tissues. Clin Chem. 1985;31;1861-1864.
Sealy J, Laragh JH. How to do plasma renin assay. In: Laragh JH, ed. Topics in Hypertension. New York, NY: Yorke Medical Books; 1980:244-256.
Bodola F, Benedict CR. Rapid simplified radioimmunoassay of arginine vasopressin and atrial natriuretic peptide in plasma. Clin Chem. 1988;34:970-973.
Little RJA, Rubin DB. Statistical Analysis and Missing Data. New York, NY: Wiley and Sons; 1987.
Searle SR. Linear Models. New York, NY: Wiley and Sons; 1971:340-361.
Konstam MA, Rousseau MF, Kronenberg MW, Udelson JE, Melin J, Stewart D, Dolan N, Edens TR, Ahn S, Kinan D, Howe DM, Kilcoyne L, Metherall J, Benedict C, Yusuf S, Pouleur H, for the SOLVD Investigators. Effects of the angiotensin converting enzyme inhibitor enalapril on the long-term progression of left ventricular dysfunction in patients with heart failure. Circulation. 1992;86:431-438.
Pouleur H, Rousseau MF, van Eyll C, Stoleru L, Hayashida W, Udelson JA, Dolan N, Kinan D, Gallagher P, Ahn S, Benedict CR, Yusuf S, Konstam K for the SOLVD Investigators. Effects of long-term enalapril therapy on left ventricular diastolic properties in patients with depressed ejection fraction. Circulation. 1993;88:481-491.
Anversa P, Loud AV, Levicky V, Guideri G. Left ventricular failure induced by myocardial infarction, II: tissue morphometry. Am J Physiol. 1985;248:H883-H889.
Katz AM. Cardiomyopathy of overload: a major determinant of prognosis in congestive heart failure. N Engl J Med. 1990;322: 100-110.
Quinones MA, Weiner DH, Shelton BJ, Greenberg BH, Limacher MC, Koilipillai C, Shindler DM, Yusuf S, for the SOLVD Investigators. Echocardiographic predictors of one-year clinical outcome in Study of Left Ventricular Dysfunction (SOLVD) Trial and Registry: an analysis of 1172 patients. Circulation. 1993;88(suppl I):I-304. Abstract.
Gunther S, Grossman W. Determinants of ventricular function in pressure-overload hypertrophy in man. Circulation. 1979;59: 679-688.
Baker KM, Aceto JF. Angiotensin II stimulation of protein synthesis and cell growth in chick heart cells. Am J Physiol. 1990;259:H610-H618.
Aceto JF, Baker KM. [Sar′]Angiotensin II receptor-mediated stimulation of protein synthesis in chick heart cells. Am Physiol Soc. 1990;27:H806-H813.
Choong CY, Abascal VA, Thomas JD, Guerrero JL, McGlew S, Weyman AE. Combined influence of ventricular loading and relaxation on the transmitral flow velocity profile in dogs measured by Doppler echocardiography. Circulation. 1988;78:672-683.
Ishida Y, Meisner JS, Tsujioka K, et al. Left ventricular filling dynamics: influence of left ventricular relaxation and left atrial pressure. Circulation. 1986;74:187-196.
Stoddard MF, Pearson AC, Kern MJ, Ratcliff J, Mrosek DG, Laboritz AJ. Influence of alteration in preload on the pattern of left ventricular diastolic filling as assessed by Doppler echocardiography in humans. Circulation. 1989;79:1226-1236.
Raine AEG, Arne P, Burgisser F, et al. Atrial natriuretic peptide and atrial pressure in patients with congestive heart failure. N Engl J Med. 1986;31:533-537.
Cody RJ, Atlas SA, Laragh JH, et al. Atrial natriuretic factor in normal subjects and heart failure patients. J Clin Invest. 1986;78: 1362-1374.