Left ventricular (LV) hypertrophy on electrocardiography or increased left ventricular mass (LVM) on echocardiography are strong predictors of death and nonfatal cardiovascular events in patients with hypertension, coronary heart disease, other cardiovascular diseases, and in the general population (1- 5). Moreover, similar to other cardiovascular risk factors, such as cholesterol and blood pressure (BP), the relationship between LVM and cardiovascular risk may extend to increases in LVM within the “normal” (average) range (5). This may be particularly important in patients with other markers of increased cardiovascular risk, such as those with preexisting cardiovascular disease or those with multiple risk factors. Similarly, increased LV volume and abnormal LV systolic function are potent predictors of events in patients with coronary heart disease (6- 8).

Angiotensin-converting enzyme (ACE) inhibitors reduce LVM and regress LV hypertrophy in animal models and in hypertensive human subjects (9- 11). Similarly, ACE inhibitors have been shown to reduce LV volumes and to improve LV systolic function in patients with coronary heart disease when administered early after myocardial infarction or in those with low left ventricular ejection fraction (LVEF) (8,12). These beneficial effects of ACE inhibitor therapy on LVM and LV remodeling may be mediated through the hemodynamic, BP and afterload-reducing actions of these agents, but also through load-independent mechanisms related to blockade of angiotensin II-mediated myocyte hypertrophy, reduction in cardiac extracellular collagen matrix formation attained by blocking angiotensin II, and by decrease in aldosterone release and through bradykinin-mediated actions (13- 15).

Patients with atherosclerotic vascular disease have increased expression of tissue renin-angiotensin systems in the vasculature and in the heart (14- 15), which could contribute to an increase in LVM and to LV dilation. We postulated that chronic ACE inhibitor therapy would prevent an increase in LVM and volumes and a decline in LV systolic function in patients with atherosclerotic vascular disease with average (“normal”) BP and LVM and with preserved LV systolic function. We further postulated that this effect would be largely independent of BP-lowering. We also sought to evaluate the relationship between ramipril dose and its effects on LVM, volume, and function.

Therefore, we evaluated the effects of ramipril on LV structure and function in an echocardiographic substudy of the Heart Outcomes Prevention Evaluation (HOPE) trial (16- 17).

Patients were recruited between December 1993 and August 1995 from three Canadian HOPE study centers, selected based on expertise and interest in the study. The study protocol was approved by the ethics board of each participating institution, and all patients provided informed consent. All of the HOPE study participants from these centers who had technically adequate baseline echocardiograms were enrolled in the echocardiographic substudy. Over 85% of patients were recruited at McMaster University, which also served as the core echocardiography laboratory. Detailed patient eligibility criteria for the HOPE trial have been published (16- 17). In summary, patients were eligible if they were ≥55 years old, had vascular disease or diabetes and at least one additional cardiovascular risk factor and did not have heart failure symptoms, known low LVEF (<40%), uncontrolled hypertension, renal disease (serum creatinine >200 μmol/l or proteinuria), or other major cardiac or noncardiac illness. To be eligible for the echocardiographic substudy, patients had to have also an adequate baseline echocardiographic examination, defined as availability of recordings of good or fair quality for M-mode and two-dimensional echocardiography, allowing measurements of at least two of the echocardiographic outcome measurements of LVM, LV volumes, and LVEF.

The echocardiographic substudy of the HOPE trial was a randomized, double-blind, placebo-controlled trial that evaluated two doses of ramipril (2.5 mg/day and 10 mg/day). Patients were also randomized to 400 IU/day of vitamin E or placebo, as part of the parent HOPE trial. The study was designed to have >80% power to detect a change of 10 g/m² in left ventricular mass index (LVMI) assuming a standard deviation of 25 g/m². Baseline echocardiograms were performed at the randomization visit. Follow-up visits occurred one month after randomization and every six months thereafter. Study-end echocardiograms were performed at the four-year visit. Blood pressure was measured by experienced study nurses at randomization, one month, two years, and study end using standard sphygmomanometers and a standardized protocol (study visits generally occurred during morning hours, use of appropriate cuff size was ensured, patients were left alone supine for ≥5 min, thereafter two exact readings avoiding rounding were obtained from each arm, and the lowest readings from the left and right arms were averaged). The study protocol did not include algorithms for the management of cardiovascular risk factors, and the study drug dose was not titrated to attain predefined BP goals. The patients' usual (clinical) physician(s) managed all cardiovascular risk factors, including BP. The study protocol and the study personnel strongly recommended adherence to contemporary guidelines for the management of cardiovascular risk factors. The open-label use of ACE inhibitors for BP management alone was not allowed. These agents could be used only when clear new changes in the patients' clinical status occurred, for which ACE inhibitor therapy is generally strongly recommended. If such clinical changes were transient, every effort was made to discontinue treatment with open-label ACE inhibitors.

Echocardiograms were acquired by trained study sonographers following a standardized protocol and using high-quality commercially available echocardiographs equipped with 3.0- to 3.5-MHz and 2.0- to 2.5-MHz probes. Standardized examinations included two-dimensional-guided M-mode recordings obtained in the parasternal long- and short-axis views below the mitral leaflet tips to the nearest millimeter and two-dimensional parasternal long- and short-axis (at the mitral leaflets, midpapillary, and apical levels) and apical (four-chamber, two-chamber, and long-axis) views. All measurements were performed off-line by one reader (R.S.) blinded to subjects' randomization status and clinical data and using a computerized review station (Freeland Systems, Cine View, Indianapolis, Indiana). A minimum of three cardiac cycles for patients in sinus rhythm and five cardiac cycles for patients in atrial fibrillation were captured and measured. The M-mode measurements included end-diastolic LV internal diameter, interventricular septal thickness and posterior wall thickness, and end-systolic and LV internal diameter. Cases where accurate M-mode measurements could not be obtained in the parasternal short axis were substituted by parasternal long-axis measurements. All M-mode measurements were performed using the American Society of Echocardiography (ASE) leading-edge convention (18). When optimal orientation of the LV imaging views could not be obtained, correctly oriented two-dimensional linear measurements were obtained using the ASE leading-edge convention. The LVM was calculated using the corrected ASE method (19), and LVMI was obtained by indexing for body surface area. Between-sonographer (on 83 duplicate examinations) and within-reader (on 80 duplicate readings) intraclass correlation coefficients for LVM were 0.84 and 0.85, respectively. Two-dimensional LVM, left ventricular end-diastolic volume (LVEDV) and left ventricular end-systolic volume (LVESV), and LVEF were calculated from two-dimensional recordings using the modified biplane Simpson's method (20). Wall motion score (WMS) and new wall motion abnormalities (NWMAB) were evaluated on two-dimensional echocardiograms using the 16-segment model (21). Between-sonographer and within-reader (on 20 duplicate readings) intraclass correlation coefficients for LVEF measurements were 0.70 and 0.75, respectively. Reproducibility of two-dimensional LVM measurements was lower than that of M-mode measurements and, therefore, the latter were used for the primary analysis of treatment effects on LVM and LVMI. An attempt was made to obtain all measurements from all echocardiograms of all 446 patients included in the final analysis. However, at times the image quality was suboptimal. Consequently, not all measurements were obtained in every patient. M-mode LVM calculations were possible in 99% of the patients; two-dimensional measurements of LVEDV, LVESV, LVEF, LVM, and WMS were obtained in 98%, 98%, 98%, 96%, and 98% of patients, respectively.

Predefined echocardiographic outcome measurements were the changes between study end and baseline in LVM, LVMI, LVEDV, LVESV, LVEF, and WMS. In addition, the effect of the study intervention on NWMAB was evaluated.

All analyses were by intention-to-treat and were done in SAS 8.0 (SAS Institute, Cary, North Carolina). Baseline characteristics were compared by one-way analysis of variance and chi-square tests as appropriate. There was no significant interaction between ramipril and vitamin E for any of the echocardiographic measurements. Therefore, analyses were done to evaluate the overall impact of ramipril on each predefined echocardiographic outcome and to compare differences between each of the two doses of ramipril versus placebo. The predefined primary analysis was by analysis of variance, with changes in LVM, LVMI, LVEDV, LVESV, WMS, and NWMAB, respectively, as the dependent variables and treatment assignment and baseline measurements of the echocardiographic parameter analyzed as the independent variables. In addition, analyses controlling for age, gender, and baseline echocardiographic measurements, for changes in systolic and in diastolic BP and with multivariate adjustment for baseline imbalances, and for other variables that may influence LVM and function (age, gender, body mass index, history of hypertension, and history of coronary heart disease) were performed by analysis of covariance. The analysis evaluating the overall ramipril effect used ramipril as a continuous variable with values of 0, 2.5 mg, and 10 mg in the model to test the linear impact of ramipril on LVM and function. Dunnett's test for comparison of multiple treatments against one control was applied in the analyses comparing each dose of ramipril versus placebo. For all analyses, the level of statistical significance was set at p < 0.05.

A total of 506 patients were enrolled in the study. The baseline characteristics were well-balanced with the exception of small differences in body mass index and in history of coronary heart disease (Table le1). A total of 446 patients completed the study, 151 in the placebo, 149 in the ramipril 2.5 mg/day, and 146 in the ramipril 10 mg/day groups. Their baseline characteristics were similar to those of the 506 patients randomized and were also well-balanced.

Clinical follow-up was complete in all study patients. Of the initial 506 study patients, 44 (9%) died, and 16 (3%) missed the study-end echocardiogram due to illness or change in residence or had inadequate study-end echocardiograms. Thus, study-end echocardiograms were obtained in 446 study participants (88% of all randomized patients and 97% of those alive at study end). Adherence and use of non-study ACE inhibitors are shown in (Table le2).

Both systolic and diastolic BP was well-controlled at baseline, 131/76 mm Hg, and similar in the three treatment arms. Mean baseline, one month, two years, and study-end BP measurements are shown in (Table le3).

Ramipril significantly reduced the change in LVM (p = 0.03 for the overall dose-dependent effect of ramipril in the analysis of variance analysis) and LVMI (p = 0.02 for the overall effect of ramipril). The benefit of ramipril therapy on LVM and LVMI remained statistically significant after adjustment for age, gender, and baseline LVM and LVMI, respectively, and after adjustment for changes in systolic and in diastolic BP (p < 0.05). In the multivariate model adjusting for baseline imbalances and predictors of LVM and function, similar trends were observed (p = 0.08 for the change in LVM and in LVMI). The treatment benefit was derived entirely from the ramipril 10 mg/day group with no significant changes for the comparison between the ramipril 2.5 mg/day versus placebo groups (Table le4). The mean differences in the changes in LVM and in LVMI between the ramipril 10 mg/day and placebo groups were 11.74 g and 6.0 g/m², respectively. The change in the sum of the interventricular septal thickness and posterior wall thickness, which reflects the thickness of the myocardium and is largely load-independent, was reduced by treatment with ramipril 10 mg/day (mean difference, 0.57 mm; p < 0.05). The main analyses of the treatment effect on changes in LVM and LVMI were derived from M-mode echocardiography. Data obtained from two-dimensional echocardiography also showed similar significant treatment benefits for ramipril. The mean differences in two-dimensional echocardiographic measurements of changes in LVM and in LVMI between the ramipril 10 mg/day and placebo groups were 14.06 g (p = 0.03) and 7.21 g/m² (p = 0.03), respectively.

The LVEDV and LVESV increased over time in the placebo arm of the trial, but decreased in the ramipril 10 mg/day group, with similar but statistically not significant trends for the ramipril 2.5 mg/day group (Table le4). The overall treatment effect of ramipril and the effect of ramipril 10 mg/day on changes in LVEDV and in LVESV remained statistically significant after adjusting for age, gender, and baseline LVEDV and LVESV, respectively, for changes in systolic and in diastolic BP and in the multivariate model. The differences in LVEDV and LVESV between the ramipril 10 mg/day and the placebo groups were 10.06 ml (p = 0.019) and 7.21 ml (p = 0.002), respectively.

Baseline LVEF was well-preserved and similar, 58%, in all three study groups. In patients assigned to the placebo arm of the trial, there was a 2% loss in LVEF over time, with only minimal 0.17% loss in LVEF in the ramipril 10 mg/day group and an intermediate 1.5% loss in LVEF in the ramipril 2.5 mg/day group (Table le3). The effect of ramipril was statistically significant (p = 0.009 for the overall dose-dependent ramipril effect adjusted for baseline LVEF, p = 0.011 in the model adjusted for age, gender, and baseline LVEF, p = 0.025 in the multivariate model, and p = 0.062 in the model adjusted for time-dependent changes in systolic and in diastolic BP). The mean difference between changes in LVEF in the ramipril 10 mg/day and placebo groups was 1.85% (p = 0.018). There were similar trends towards better WMS and fewer NWMAB in the ramipril 10 mg/day group, although these did not reach statistical significance (Table le4).

In order to better understand the observed treatment benefit on LVEF, we evaluated the effects of ramipril in patients who did and those who did not sustain a myocardial infarction during the study. Of the 446 patients who completed both echocardiographic examinations, 39 sustained a myocardial infarction (16 in the placebo, 12 in the ramipril 2.5 mg/day, and 11 in the ramipril 10 mg/day groups), and their LVEF changes did not differ significantly, probably due to the small number of observations and the limited power of this analysis. Among the 407 patients who did not sustain a myocardial infarction during the study, there was a significant treatment effect (p = 0.028 for the overall dose-dependent ramipril effect and p = 0.019 for the comparison between the ramipril 10 mg/day and the placebo groups). These differences remained statistically significant in the model adjusted for age, gender, and baseline LVEF and in the multivariate model, with a strong trend (p = 0.06) in the analysis corrected for systolic and for diastolic BP changes.

As expected in this relatively small substudy, there were no statistically significant differences in the primary clinical outcome (the composite of cardiovascular death, myocardial infarction, and stroke) that occurred in 32 patients (19.1%) in the placebo, 24 (14%) in the ramipril 2.5 mg/day, and 25 (15%) in the ramipril 10 mg/day groups. Reliable data on clinical outcomes are provided by the adequately powered parent HOPE trial (17).

We demonstrated that long-term therapy with the ACE inhibitor ramipril has favorable effects on LV structure and function, by reducing LVM, LV volumes, and the loss in LVEF in high-risk patients with vascular disease, with well-controlled BP, and with average (“normal”) baseline LVM, LV volume, and LVEF. These are novel observations, as previous clinical ACE inhibitor studies have selected patients based on elevated BP and/or increased LVM or depressed LV systolic function. Furthermore, in such previous trials, ACE inhibitor dose was generally titrated based on BP response.

Both ramipril 10 mg/day and 2.5 mg/day had significant, but modest, BP-lowering effects at one month and at two years after randomization although, by study end, systolic BP returned to baseline levels in the active treatment groups and increased in the placebo group. The magnitude of BP-lowering was similar for the two ramipril doses, which may be related to the fact that most study patients were taking various other anti-ischemic drugs at different doses, not controlled by the study protocol. By contrast, the benefits of ramipril on LVM, LV volumes, and LV function were observed with the higher dose of 10 mg/day, while the lower dose of 2.5 mg/day did not have a significant impact. These observations suggest that the benefits of ACE inhibitor therapy on LVM and function are, at least in part, independent of BP-lowering. Furthermore, statistical analyses adjusting for BP changes did not significantly alter the study results for the analyses of treatment effects on LVM and LV volumes, while for LVEF a strong trend towards treatment benefit remained after correcting for changes in BP (p = 0.06), and the multivariate model, which included history of hypertension, remained statistically significant. In the multivariate model, which adjusted for baseline LVM, age, gender, and for baseline imbalances and other predictors of LVM, there was a strong trend for an overall ramipril effect on LVM and LVMI (p = 0.08). While this analysis did not reach traditional levels of statistical significance, a strong trend was observed; this analysis is conservative and controls not only for parameters that differed at baseline, and, furthermore, the comparison between the ramipril 10 mg/day and the placebo groups was significant in the multivariate model.

The beneficial effect of ramipril on LVEF was not restricted to patients who had an interim infarct during the study. This analysis suggests that the impact of ramipril on preservation of LV systolic function is likely multifactorial and cannot be explained by prevention of new infarcts or reduction in BP alone.

Previous studies in hypertensive patients have reported weak correlations between changes in LVM and drug-induced BP reductions, and it was suggested that LVM changes more closely relate to changes in 24-h than to “office” BP and that clinical trials may underestimate the impact of intrapatient variability in BP reduction on LVM changes (22- 23). However, these studies have studied patients selected based on elevated baseline BP, while in our trial a history of hypertension was present in only 33% of the study participants, and BP was, on average, well-controlled, 131/76 mm Hg.

The mean difference in change over four years in LVMI between the placebo and ramipril 10 mg/day groups was 6 g/m², which is less than observed in some BP-lowering trials (22). However, such trials have selected patients who were hypertensive and had baseline LV hypertrophy, while, in our study, patient selection was based on cardiovascular risk alone, and most patients had average (“normal”) baseline BP and LVM.

Our findings are concordant with other analyses of the HOPE trial and with previous investigations. Thus, in the HOPE trial, ramipril caused prevention or regression of LV hypertrophy defined on electrocardiography, an effect that was largely independent of BP changes (24). Other clinical trials have also shown regression of LV hypertrophy with ACE inhibitors in hypertensive populations and some (10- 11,25), although not all such trials (22), suggest that, for similar degrees of BP-lowering, these agents may be more effective than other antihypertensive agents in reducing LV hypertrophy.

The mechanism of reduction in LVM with ACE inhibitor therapy is likely multifactorial. Angiotensin II is a potent vasoconstrictor, has positive inotropic actions on the heart muscle, and can further affect cardiac hemodynamics by potentiating the actions of the sympathetic nervous system (13- 14). In addition, both circulating and locally produced angiotensin II has a direct trophic effect on myocardial smooth muscle cells and contributes to restructuring and proliferation of the myocardial extracellular collagen matrix, thus promoting LV hypertrophy. The ACE inhibitors can prevent or regress LV hypertrophy through hemodynamic afterload reducing actions, but also through load-independent mechanisms, as shown in animal models (13- 14,26- 27).

The reduction in LV volume and the preservation of LV systolic function observed in our study are also consistent with the reduced rates of heart failure in the parent HOPE trial (28) and with previous studies (8,12) and is related to the hemodynamic afterload-reducing effect of ramipril, but also to the prevention of new infarcts, to the benefit on myocardial remodeling, and to the general stabilizing effect of ACE inhibition on atherosclerosis (13,15). As shown in experimental studies, these effects are related to blockade of angiotensin II type 1 receptor-mediated actions, but also to increased availability of bradykinin and possibly additional ACE inhibitor effects (13- 15,29).

In summary, our study shows a significant dose-dependent, but, at least in part, BP-independent beneficial effect of ACE inhibitor therapy on LV structure and function. These results parallel similar findings on the prevention of clinical events with ramipril observed in the parent HOPE trial, which was only partly explained by BP reduction. These data provide further evidence supporting the use of long-term ACE inhibitor therapy in a wide range of patients with cardiovascular disease.