OUP user menu

Effect of long-term ACE-inhibitor therapy in elderly vascular disease patients

Monica Gianni, Jackie Bosch, Janice Pogue, Jeffrey Probstfield, Gilles Dagenais, Salim Yusuf, Eva Lonn
DOI: http://dx.doi.org/10.1093/eurheartj/ehm017 1382-1388 First published online: 29 March 2007


Aims Cardiovascular (CV) disease is the leading cause of death in the elderly. The use of ACE-inhibitors in elderly patients with chronic stable vascular disease has not been previously reported.

Methods and results The HOPE trial evaluated the effects of ramipril and vitamin E in high-risk vascular disease patients. We report the effects of ramipril in the elderly HOPE study patients, defined as those ≥70 years of age. A total of 2755 elderly patients with vascular disease or diabetes and at least one additional CV risk factor and without heart failure or low ejection fraction were randomized to ramipril 10 mg daily or placebo. Those assigned to ramipril had fewer major vascular events compared to those assigned to placebo [18.6 vs. 24.0%, hazard ratio (HR) = 0.75, P = 0.0006], CV deaths (9.3 vs. 13.0%, HR = 0.71, P = 0.003), myocardial infarctions (12.0 vs. 15.6%, HR = 0.75, P = 0.006), and strokes (5.4 vs. 7.7%, HR = 0.69, P = 0.013). Treatment was safe and generally well tolerated.

Conclusion Ramipril reduces the risk of major vascular events in elderly patients with vascular disease and is safe and well tolerated by most.

  • Cardiovascular disease
  • Myocardial infarction
  • Stroke
  • Age
  • ACE-inhibitor
See page 1277 for the editorial comment on this article (doi:10.1093/eurheartj/ehm017)


Cardiovascular (CV) diseases account for about 40–50% of all deaths in elderly people in western countries.1 CV risk increases sharply with age and recent guidelines recommend aggressive targeting of aging people for CV risk reduction.2 However, the elderly have often been excluded from CV randomized controlled trials and few studies have evaluated the effects of various drugs in secondary prevention in these patients.3 Coupled with the resultant paucity of data on efficacy and safety of CV drugs in elderly patients, there is concern related to their perceived higher risk of adverse drug reactions. Indeed, elderly subjects have a higher incidence of autonomic dysfunction and are more prone to orthostatic hypotension; their kidney and liver functions are often impaired resulting in reduced excretion rates of drug metabolites and they frequently require drugs for several concomitant illnesses and are thus at risk for drug interactions. Therefore, physicians are often reluctant to use CV preventive drugs in elderly patients.

Angiotensin-Converting Enzyme (ACE) inhibitors have been studied in elderly patients with hypertension, stroke, heart failure, and diabetes.46 However, to-date no studies have specifically evaluated the use of ACE-inhibitors in elderly patients with chronic stable vascular disease.

The Heart Outcomes Prevention Evaluation (HOPE) trial showed reduced CV events in high-risk patients with stable vascular disease or with diabetes treated with ramipril.7 We report the efficacy and safety of ramipril in the elderly HOPE patients, those in their 7th and 8th decade of life.


HOPE was a randomized clinical trial, which evaluated the high-tissue affinity ACE-inhibitor ramipril in 9297 high-risk patients with stable vascular disease or diabetes. Detailed descriptions of the HOPE study design have been published.7,8 The current report is a post hoc analysis of the effects of ramipril in elderly patients.

The study complies with the Declaration of Helsinki, was approved by the ethics boards of all participating centres, was monitored by an independent Data Safety and Monitoring Board (DSMB) and all patients provided written informed consent. The authors of this report, the Data Management and Statistical Center and the DSMB were fully independent of the study sponsors.

Study population

Men and women ≥55 years with a history of coronary artery disease (CAD), stroke, peripheral arterial disease (PAD), or with diabetes plus at least one other CV risk factor were enrolled. There was no upper age limit for eligibility. Patients were excluded if they had known heart failure or low left ventricular (LV) ejection fraction (<40%), uncontrolled hypertension, renal artery stenosis, serum creatinine ≥200 µmol/L, overt nephropathy, hyperkalemia (serum potassium >5.5 mEq/L), myocardial infarction (MI), unstable angina, or stroke within 4 weeks of enrolment, use of or intolerance to ACE-inhibitors or vitamin E or other major life-threatening illnesses. Recruitment occurred from December 1993 to June 1995 in Canada, the USA, Western Europe, and South America. In this post hoc analysis we defined study patients aged ≥ 70 years as ‘elderly’.

Study design

After a single-blind run-in with ramipril 2.5 mg for 7–10 days followed by placebo for 10 to 14 days, compliant patients who did not develop side effects were assigned by central concealed randomization to ramipril 10 mg daily or matching placebo and to natural source Vitamin E 400 IU daily or matching placebo in a 2 × 2 factorial design. Ramipril was titrated to 10 mg/day over 1 month, using 2.5 mg/day for the first week and 5 mg/day for 3 weeks thereafter. Median follow-up was 4.4 years (interquartile range, 4.0–4.6 years).

Follow-up and study outcomes

The primary outcome was the composite of CV death, MI, and stroke. Each of the components of this outcome was analysed separately. Secondary outcomes were all-cause death, revascularizations, and hospitalizations for unstable angina or heart failure, nephropathy and complications related to diabetes. Additional outcomes included all episodes of heart failure and the development of diabetes. Detailed definitions of these outcomes have been published.7

Statistical analysis

Baseline characteristics were compared using t-tests or chi-square tests, as appropriate. Survival curves were estimated according to the Kaplan–Meier procedure, and treatments were compared by the log-rank test. All analyses were performed according to the intention-to-treat principle and were stratified for treatment allocation to Vitamin E or placebo to account for the factorial study design. Possible interactions between ramipril and Vitamin E were evaluated. Cox proportional hazard models were fitted to examine the effect of age using models with a treatment allocation (ramipril or placebo) × age (age groups were defined as those <70 and those ≥70 years) interaction term (the assumption of proportional hazards was examined graphically and no evidence of violation was found). Subgroup analyses evaluated the consistency of the effect of ramipril on the primary study outcome among subsets of elderly patients. The effect of ramipril on blood pressure (BP) by age group was examined in regression analyses using repeated measures mixed models with compound symmetry with the change in systolic and in diastolic BP as dependent variables. All statistical tests were two-sided and the level of significance was set at 2α = 0.05.


Baseline characteristics

A total of 10 576 patients, 3007 ≥ 70 years and 7569 < 70 years entered the run-in phase. Of these 1035, 190 (6.3%) ≥70 years and 845(11.2%) <70 years were subsequently excluded from randomization because of non-compliance (<80% of pills taken) reported in 2.6% of patients ≥70 years and 4.2% of those <70 years, withdrawal of consent, 3.0% and 4.1%, respectively, or side effects. Among reported side effects, abnormal creatinine (>250 µmol/L) occurred in six (0.2%) patients ≥70 years and seven (0.1%) <70 years, hyperkalemia (>5.5 mEq/L) in seven (0.2%) and 41 (0.5%), cough in nine (0.3%) and 30 (0.4%), hypotension or dizziness in eight (0.3%) and 45 (0.6%) and angioedema in zero and five (0.1%), respectively. Of the remaining 9541, 244 (62 ≥ 70 years and 182 < 70 years) were randomly assigned to receive a lower dose (2.5 mg per day) of ramipril or placebo as part of a substudy and are not part of the main study analysis (addition of these patients does not substantially alter the study results). Serum creatinine measured after the active ramipril phase of the run-in period was 101.5 µmol/L ± 23.8 in patients ≥70 years and 95.0 µmol/L ± 20.2 in those <70 years (P < 0.0001).

There were 2755 patients aged ≥ 70 years and 6542 < 70 years randomized to ramipril 10 mg or placebo. Among the elderly HOPE study patients, 1757 were 70–74 years, 738 were 75–79 years, 230 were 80–84 years, and 30 were ≥85 years. Most elderly HOPE study patients had a history of CAD (83.1%) and 53.3% reported a prior MI. There was a high prevalence of CV risk factors, including hypertension (47.8%), diabetes (34.5%), hypercholesterolemia (58.8%), and microalbuminuria (26.3%), although few elderly patients were current smokers (7.6%) (Table 1). Compared to patients <70 years of age, those ≥70 years, had higher systolic BP, were more likely to be a female and to have a history of PAD, LV hypertrophy on electrocardiogram, and microalbuminuria. Current smoking, diabetes, and hypercholesterolemia were more common in patients <70 years. Fewer patients ≥70 years were taking lipid-lowering drugs, but more were taking diuretics and calcium channel blockers. Baseline characteristics did not differ significantly among patients ≥70 years in the ramipril and placebo groups (Table 1).

View this table:
Table 1

Baseline characteristics of the HOPE study patients ≥70 years and <70 years

CharacteristicsAge ≥70 years (n = 2755)Age <70 years (n = 6542)PaAge ≥70 years ramipril (n = 1375)Age ≥70 years placebo (n = 1380)
Age (years)74.1 ± 3.762.4 ± 4.3<0.000174.0 ± 3.674.2 ± 3.7
Systolic BP (mmHg)142.2 ± 20.3137.2 ± 19.2<0.0001141.8 ± 20.4142.6 ± 20.2
Diastolic BP (mmHg)77.5 ± 10.679.5 ± 10.5<0.000177.6 ± 10.777.4 ± 10.5
Heart rate (b.p.m./min)68.2 ± 11.468.8 ± 11.30.01268.2 ± 11.368.3 ± 11.4
Body mass index (Kg/m2)26.8 ± 4.128.1 ± 4.4<0.000126.9 ± 4.226.7 ± 3.9
Female gender (%)830 (30.1%)1650 (25.2%)<0.0001435 (31.6%)395 (28.6%)
History of
 Coronary artery disease2290 (83.1%)5187 (79.3%)<0.00011137 (82.7%)1153 (83.6%)
 MI1468 (53.3%)3361 (51.4%)<0.4729 (53.0%)739 (53.6%)
 Stable angina1602 (58.1%)3560 (54.4%)0.001785 (57.1%)817 (59.2%)
 Unstable angina743 (27.0%)§1624 (28.8%)0.03380 (27.6%)363 (26.3%)
 CABG surgery763 (27.7%)1636 (25.0%)0.007371 (27.0%)392 (28.4%)
 PCI409 (14.8%)1250 (19.1%)<0.0001205 (14.9%)204 (14.8%)
 Stroke or TIA401 (14.5%)612 (9.3%)<0.0001186 (13.5%)215 (15.6%)
 Peripheral artery diseaseb1307 (47.4%)2521 (38.5%)<0.0001645 (46.9%)662 (48.0%)
 Hypertension1317 (47.8%)3038 (46.4%)0.23652 (47.4%)665 (48.2%)
 Diabetes950 (34.5%)2627 (41.6%)<0.0001448 (32.6%)502 (36.4%)
 Cholesterol >5.2 mmol/L1621 (58.8%)4504 (68.8%)<0.0001818 (59.5%)803 (58.2%)
 HDL-C <0.9 mmol/L443 (16.1%)1280 (19.6%)<0.0001219 (15.9%)224 (16.2%)
 Current cigarette smoking209 (7.6%)1110 (16.9%)<0.0001110 (8.0%)99 (7.2%)
 LV hypertrophy279 (10.1%)506 (7.7%)0.0002134 (9.7%)145 (10.5%)
 Microalbuminuria726 (26.3%)1237 (18.9%)<0.0001341 (24.8%)385 (27.9%)
 Aspirin2111 (76.6%)4963 (75.9%)0.431035 (75.3%)1076 (78.0%)
 β-blockers1066 (38.7%)2607 (39.8%)0.30515 (37.5%)551 (39.9%)
 Calcium channel blockers1378 (50.0%)3002 (45.9%)0.0003678 (49.3%)700 (50.7%)
 Diuretics530 (19.2%)889 (13.6%)<0.0001258 (18.8%)272 (19.7%)
 Lipid-lowering drugs615 (22.32%)2043 (31.2%)<0.0001319 (23.2%)296 (21.4%)
  • Data are presented as the mean ± SD or number (%) of subjects.

  • aFor comparisons between study patients aged ≥70 years and those <70 years.

  • bPeripheral artery disease was defined as history of claudication, or >50% stenosis by angiography, previous surgery or angioplasty, or ankle/arm BP ratio <0.90.

  • MI, myocardial infarction; CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention; TIA, transient ischaemic attack; HDL-C, HDL-Cholesterol.

Follow-up and adherence

In all but one patient aged ≥70 years, vital status was ascertained at the end of the study. In the active ramipril group, 85.5% of eligible patients were taking study drug or open label ramipril at 1 year, 84.2% at 2 years, 80.6% at 3 years, 73.1% at 4 years, and 75.8% at study end. In the placebo group, 87.5% were taking study drug at 1 year, 83.2% at 2 years, 77.3% at 3 years, 69.5% at 4 years, and 70.7% at study end and 12.1% were on open label ramipril at study end. The corresponding rates for patients aged <70 years were 88.2, 85.3, 82.9, 76.0, and 80.1% in the active ramipril group and 90.0, 84.7, 80.4, 71.3, and 73.5%, respectively, in the placebo group, with 12.4% taking open label ramipril at study end. Among patients on study drug, the median daily dose was 10 mg, similar in patients ≥70 years and <70 years (89% of patients ≥70 years and 90% of patients <70 years on study drug were taking 10 mg daily). Overall 35.9% of patients aged ≥70 years and 31.1% of patients aged <70 years in the active ramipril group stopped study drug at any time (P = 0.01); among patients in the placebo group 32.7% aged ≥70 years and 29.9% <70 years stopped study drug (P = 0.06). In a regression model with study drug discontinuation as the outcome variable, the treatment allocation × age group interaction term did not predict a population more likely to discontinue study drug (P = 0.36). These findings suggest slightly lower overall adherence rates among elderly patients, but no significant differences in adherence between elderly and younger patients based on ramipril allocation. Use of non-study ACE-inhibitors was generally low and similar in elderly and younger patients and was commonly related to the occurrence of clinical events judged to warrant open label ACE-inhibitor therapy.

Ramipril was generally equally well tolerated in patients aged ≥70 years and those <70 years. The most common reasons for study drug discontinuation were patients’ refusal to continue study drug, the occurrence of clinical events (often mandating the open label use of ACE-inhibitor therapy), cough, physicians’ advice, and hypertension (Table 2). When compared with patients aged <70 years, those aged ≥70 years were more likely to discontinue ramipril due to refusal to continue study medication, occurrence of clinical events, and hypotension or dizziness. However, among those aged ≥70 years only cough, angioedema and patient refusal to continue study drug were more common in the ramipril group than in the placebo group. Hypotension was more common in patients ≥70 years, but similar in those assigned to ramipril and to placebo. Renal dysfunction and hyperkalemia were relatively rare and occurred at similar rates in those older and younger than 70 years and in those in the ramipril and placebo groups, respectively.

View this table:
Table 2

Reasons for discontinuation of study drug*

Reason for stopping study drugaAge ≥70 years (n = 2755)Age <70 years (n = 6542)
Ramipril group (n = 1375)Placebo group (n = 1380)Ramipril group (n = 3270)Placebo group (n = 3272)
Patient refusal to continue study drug240 (17.5%)206 (14.9%)*454 (13.9%)425 (13.0%)
Clinical event127 (9.2%)154 (11.2%)*179 (5.5%)262 (8.0%)*
Cough101 (7.3%)21 (1.5%)*239 (7.3%)64 (2.0%)*
Hypertension33 (2.4%)48 (3.5%)76 (2.3%)135 (4.1%)
Physician advice54 (3.9%)52 (3.8%)107 (3.3%)104 (3.2%)
Hypotension or dizziness36 (2.6%)32 (2.3%)52 (1.6%)38 (1.2%)
Renal dysfunction22 (1.6%)26 (1.9%)37 (1.1%)64 (2.0%)
Hyperkalemia5 (0.4%)2 (0.1%)3 (0.1%)4 (0.1%)
Angioedema5 (0.4%)0 (0%)*10 (0.3%)6 (0.2%)
  • In the active ramipril group, as compared patients <70 years, those ≥70 years were more likely to discontinue study drug due to occurrence of clinical events (HR 1.69, 95% CI, 1.36–2.10; P < 0.0001), patient refusal (HR 1.28, 95% CI, 1.11–1.48; P = 0.001), and hypotension or dizziness (HR 1.65, 95% CI, 1.08–2.51; P = 0.02). When pooling the clinically most significant side effects of ACE-inhibitor therapy, namely cough, hypotension, renal dysfunction, hyperkalemia, and angioedema, these were reported as reasons for drug discontinuation in 125 (9.1%) patients ≥70 years and in 283 (8.7%) patients <70 years (HR 1.05, 95% CI, 0.86–1.28, P = 0.63).

  • Among patients aged ≥70 years cough, angioedema and patient refusal were more common reasons for drug discontinuation in the ramipril group when compared with the placebo group, while clinical events were more common in the placebo group.

  • aThe categories are not mutually exclusive.

  • *P < 0.05; refers to comparisons between the ramipril and placebo groups.

Blood pressure

Baseline systolic BP was higher and diastolic BP was lower in patients ≥70 years than in those <70 years and changes in BP during the study were larger in elderly patients (Table 3). However, the effect of ramipril on BP was not dependent on age group, as the BP changes were larger both in the ramipril and also in the placebo groups in patients ≥70 vs. those <70 years and the treatment allocation × age group and the treatment allocation × age group × time interaction terms were not independent predictors of systolic or diastolic BP changes in regression models.

View this table:
Table 3

Baseline and follow-up BP of the HOPE study patients aged ≥70 years and <70 years

Treatment allocationBaseline1 month2 yearsStudy end
Age ≥70 years
 Systolic BP
 Ramipriln = 1375141.8 ± 20.4n = 1351135.4 ± 19.8n = 1195137.3 ± 20.7n = 1000138.2 ± 20.2
 Placebon = 1380142.6 ± 20.2n = 1358139.8 ± 19.9n = 1211140.7 ± 20.0n = 974140.2 ± 20.0
 Diastolic BP
 Ramipriln = 137577.6 ± 10.7n = 135174.7 ± 10.9n = 119474.3 ± 11.0n = 100074.0 ± 10.7
 Placebon = 138077.4 ± 10.5n = 135876.7 ± 10.9n = 121175.8 ± 10.6n = 97475.0 ± 10.6
Age <70 years
 Systolic BP
 Ramipriln = 3270137.1 ± 19.2n = 3229132.0 ± 18.8n = 2976134.2 ± 19.7n = 2683135.4 ± 18.8
 Placebon = 3272137.3 ± 19.1n = 3231136.0 ± 18.7n = 2993137.6 ± 19.1n = 2658138.1 ± 19.3
 Diastolic BP
 Ramipriln = 327079.5 ± 10.5n = 323176.8 ± 10.3n = 297676.8 ± 10.4n = 268276.5 ± 10.4
 Placebon = 327279.5 ± 10.5n = 322879.0 ± 10.6n = 299278.6 ± 10.2n = 265877.6 ± 10.7
  • Data are presented as mean ± SD.

Outcomes by treatment assignment

Ramipril significantly reduced the primary outcome and each of its components in patients aged ≥70 years (Table 4 and Figure 1). Among the secondary and other outcomes, all-cause death, revascularization, and overt nephropathy were significantly reduced and similar trends were noted for heart failure, diabetic complications, and new diagnosis of diabetes. There was no treatment effect on hospitalizations for unstable angina. The effect of ramipril was not significantly different in patients aged ≥70 years and those <70 years for any study outcomes (Table 4). Vitamin E had no significant effects and there was no interaction between ramipril and vitamin on any of the outcomes analysed.

Figure 1

Kaplan–Meier estimates for the primary outcome in the ramipril and placebo groups in patients aged ≥70 years and in those aged <70 years.

View this table:
Table 4

Effects of ramipril on primary, secondary, and other outcomes in patients ≥70 years and <70 years

Age ≥70 years (n = 2755)Age <70 (n = 6542)
Ramipril (n = 1375) n (%)Placebo (n = 1380) n (%)HR (95% CI)Ramipril (n = 3270) n (%)Placebo (n = 3272) n (%)HR (95% CI)P-value for Interactiona
Primary outcome
MI, stroke, CV death256 (18.6%)331 (24.0%)0.75 (0.64–0.88)*395 (12.1%)495 (15.1%)0.79 (0.69–0.90)*0.6737
 MIb165 (12.0%)215 (15.6%)0.75 (0.62–0.92)**294 (9.0%)355 (10.8%)0.82 (0.70–0.96)**0.5200
 Strokeb74 (5.4%)106 (7.7%)0.69 (0.51–0.92)***82 (2.5%)120 (3.7%)0.68 (0.51–0.90)***0.9534
 CV deathb128 (9.3%)179 (13.0%)0.71 (0.56–0.89)**154 (4.7%)198 (6.1%)0.77 (0.63–0.96)0.9985
Secondary outcomes
All-cause death218 (15.9%)263 (19.1%)0.82 (0.68–0.98)***264 (8.1%)306 (9.4%)0.86 (0.73–1.01)0.6897
Revascularization191 (13.9%)250 (18.1%)0.74 (0.61–0.89)*552 (16.9%)604 (18.5%)0.90 (0.80–1.01)0.0806
All heart failure191 (13.9%)217 (15.7%)0.87 (0.71–1.05)226 (6.91%)317 (9.7%)0.70 (0.60–0.83)*0.1178
Hospitalization for heart failure70 (5.1%)79 (5.7%)0.88 (0.64–1.21)71 (2.2%)82 (2.5%)0.86 (0.63–1.19)0.9566
Hospitalization for unstable angina181 (13.2%)189 (13.7%)0.95 (0.77–1.16)373 (11.4%)378 (11.6%)0.99 (0.85–1.14)0.7385
Overt nephropathy36 (2.6%)53 (3.8%)0.64 (0.42–0.98)***108 (3.3%)132 (4.0%)0.84 (0.65–1.1)0.2623
Diabetic complications13 (1.4%)20 (2.3%)0.57 (0.29–1.16)23 (1.2%)45 (2.2%)0.55 (0.33–0.91)***0.9301
Other outcomes
New diagnosis of diabetes24 (2.6%)33 (3.8%)0.68 (0.40–1.15)78 (4.1%)122 (6.1%)0.66 (0.50–0.88)**0.9460
  • aRefers to the treatment effect x age group interaction.

  • bNot mutually exclusive.

  • *P < 0.001 (refers to comparisons between the ramipril and placebo groups).

  • **P ≤ 0.01 (refers to comparisons between the ramipril and placebo groups).

  • ***P < 0.05 (refers to comparisons between the ramipril and placebo groups).

Similar reductions in risk were observed in subgroups of patients ≥70 years (Figure 2).

Figure 2

Effects of ramipril on the primary study outcome in key subgroups of patients aged ≥70 years. The dotted line represents the average treatment effect. The size of each box is proportional to the number of patients in each individual analysis. ASA, acetyl-salicylic acid; BB, beta-blocker; CVD, cardiovascular disease; DM, diabetes mellitus; Renal, renal dysfunction defined as serum creatinine ≥124 µmol/L; HTN, hypertension; LL, lowering lipid drug.

Findings were similar when the elderly population was defined more stringently as patients ≥75 years (data available on request).


To our knowledge, this is the first detailed report of the effects of long-term ACE-inhibitor therapy in elderly patients with stable vascular disease. Our study shows that treatment with ramipril 10 mg daily over 4.5 years reduces major CV events in these patients and is generally safe and well tolerated. We observed large proportional reductions in the risk for major vascular events, for MI, stroke, CV death, and all-cause death of 25, 25, 31, 25, and 18%, respectively. These were of similar magnitude to the proportional reductions in risk observed in patients aged <70 years.

Major CV event and death rates were high in the elderly HOPE study patients, in spite of including only stable patients, without recent acute ischaemic events and without heart failure or LV dysfunction; over the average 4.5 years of follow-up, 24% of patients ≥70 years in the placebo arm had a primary event, 13% died from CV causes, 16% sustained an MI, 8% had a stroke, and 19% died. These findings confirm the major role of age as a CV risk factor and support the recommendations for aggressive preventive therapies in elderly patients. Importantly, due to the high risk of CV events in elderly patients, the absolute risk reductions attained with ramipril were higher than in patients aged ≥70 years vs. those <70 years for most endpoints. For example, the absolute risk reduction for the primary endpoint was 5.4% in patients ≥70 years and 3% for those <70 years, so that for elderly patients the number needed to treat (NNT) to prevent one major CV event over 4.5 years was 18, compared to 33 for younger patients.

Our findings are consistent with most large ACE-inhibitor studies in patients with and without LV dysfunction,8,9 although previous studies have not evaluated elderly patients in detail. More recently, the PEACE trial failed to demonstrate clear benefit of trandolapril in over 8000 patients with stable CAD, including 912 aged ≥75 years.10 The results of this trial may be related to its less robust primary outcome, lower power, and possibly, the play of chance. Overall, the totality of the evidence does support the use of ACE-inhibitors in secondary prevention.9

The benefits attained with ramipril are likely related in addition to BP lowering to a wide range of cardiac and vascular protective actions,11,12 including its effects on atherosclerosis,13 LV mass and function,14 insulin sensitivity,15 and renal function.16

Ramipril was generally safe and well tolerated in most elderly patients. Angioedema, hypotension, renal dysfunction, and hyperkalemia, the most hazardous side effects of ACE-inhibitors, were rare. Patients did undergo a run-in phase, ramipril was titrated over a 4-week period and clinic visits occurred every 6 months, although protocol mandated follow-up assessments were simple, based primarily on history, without extensive laboratory testing. This suggests that with a prudent approach to patient selection and drug titration, ramipril can be used safely in most elderly vascular disease patients. Considering their high risk for CV events and the substantial risk reductions attained, ramipril had a very favourable risk to benefit ratio in these patients.

At baseline use of lipid-lowering drugs, proven to benefit vascular disease patients, was substantially lower in the elderly patients. A similar pattern of lower use of proven pharmacological interventions in elderly patients is observed in large registries17,18 and confirms physicians’ uncertainties and biases when managing elderly patients, underscoring the need for robust evidence from randomized clinical trials, that specifically address the role of CV therapies in this highly relevant and prevalent patient group. This needs to become a priority in clinical research.

Limitations of this report include the post hoc analysis and the potential threat to generalizability to the community setting, inherent to any clinical trial, especially when an active run-in period is used. However, all data were collected prospectively, all relevant efficacy and safety outcomes are reported and generalizability is high due to the wide eligibility criteria and the simple monitoring procedures used, which mimic a prudent, yet, sensible and commonly used clinical approach.

In conclusion, ramipril substantially reduces the risk of CV events and death in elderly patients with vascular disease and diabetes and is generally safe and well tolerated. Therefore, it should be used as first line therapy in the secondary prevention of CV diseases in elderly patients.


The study was partially funded by the Medical Research Council of Canada, Hoechst-Marion Roussel, Astra Zeneca, King Pharmaceuticals, Natural Source Vitamin E Association, and Negma.

Conflict of interest: all authors have received research support or travel or speaking honoraria from pharmaceutical companies involved in the manufacturing and/or marketing of Ramipril.


View Abstract