sign in sign up
<<
Home , Enzyme inhibitor, Statin

Journal of Human (2004) 18, 781–788 & 2004 Nature Publishing Group All rights reserved 0950-9240/04 $30.00 www.nature.com/jhh ORIGINAL ARTICLE Effect of and ACE inhibitors alone and in combination on clinical outcome in patients with coronary

VG Athyros1, DP Mikhailidis3, AA Papageorgiou2, VI Bouloukos1, AN Pehlivanidis1, AN Symeonidis4 and M Elisaf5, for the GREACE Study Collaborative Group 1Atherosclerosis Unit, Aristotelian University, Hippocration Hospital, Thessaloniki, Greece; 2Department of Clinical , Royal Free Hospital, Royal Free and University College Medical School, Pond Street, London, UK; 32nd Propedeutic Department of Internal Medicine, Aristotelian University, Hippocration Hospital, Thessaloniki, Greece; 4Greek Society of General Practitioners, Thessaloniki, Greece; 5Department of Internal Medicine, Medical School, University of Ioannina, Greece

We assessed the ‘synergy’ of statins and angiotensin- point in group A was 31%, (95% CI À48 to À6%, P ¼ 0.01) converting inhibitors (ACEI) in reducing vascu- in comparison to group B, 59% (95% CI À72 to À48%, lar events in patients with coronary heart disease (CHD). Po0.0001) to group C and 63% (95% CI À74 to À51%, The GREek and CHD Evaluation (GREACE) Po0.0001) to group D. There was no significant Study, suggested that aggressive reduction of low difference in RRR between groups C and D (9%, CI density lipoprotein to 2.59 mmol/l À27–10%, P ¼ 0.1). Other factors (eg the pressure) (o100 mg/dl) significantly reduces morbidity and mor- that can influence clinical outcome did not differ tality in CHD patients, in comparison to undertreated significantly between the four treatment groups. In patients. In this post hoc analysis of GREACE the conclusion, the þ ACEI combination reduces patients (n ¼ 1600) were divided into four groups cardiovascular events more than a statin alone and according to long-term treatment: Group A (n ¼ 460 considerably more than an ACEI alone. Aggressive statin þ ACEI), B (n ¼ 420; statin, no ACEI), C (n ¼ 371;no statin use in the absence of an ACEI also substantially 371;no statin, on ACEI), and D (n ¼ 349; no statin, no reduced cardiovascular events. Treatment with an ACEI ACEI). Analysis of variance was used to assess in the absence of a statin use reduced clinical events in differences in the relative risk reduction (RRR) in ‘all comparison to patients not treated with an ACEI but not events’ (primary end point) between groups. During the significantly, at least in these small groups of patients. 3-year follow-up there were 292 cardiovascular events; Journal of Human Hypertension (2004) 18, 781–788. 45 (10% of patients) in group A, 61 (14.5%) in group B, 91 doi:10.1038/sj.jhh.1001748 in group C (24.5%) and 95 events in group D (27%). The Published online 1 July 2004 RRR (95% confidence interval (CI) in the primary end

Keywords: ACE inhibitors; statins; and coronary heart disease

Introduction theless, whether combination treatment with statins and ACEIs can increase clinical benefit in CHD It has been clearly demonstrated that cholesterol- patients more than each alone has not yet been lowering with statins reduces morbidity and mor- addressed by an end point trial. tality in patients with coronary heart disease 1–4 The GREek Atorvastatin and Coronary-heart-dis- (CHD). Recently, the EURopean trial On reduction ease Evaluation (GREACE) study6–9 was a prospec- of cardiac events with in stable coronary tive, randomised, target based, open-label and Artery disease (EUROPA) showed that perindopril, intention-to-treat secondary CHD prevention trial. an angiotensin-converting (ACEI), GREACE showed that the structured management of significantly improved outcome in patients with 5 dyslipidaemia with dose titration of atorvastatin can stable CHD without apparent heart failure. Never- achieve the National Cholesterol Educational Pro- gram (NCEP) low-density lipoprotein cholesterol (LDL-C) treatment goal (o2.6 mmol/l;100 mg/dl).10 Correspondence: Dr VG Athyros, Unit, This was associated with significant reductions in 15 Marmara St, Thessaloniki 551 32, Greece. E-mail: [email protected] morbidity and mortality, in comparison to usual care. Received 19 January 2004; revised 03 April 2004; accepted 14 In the present post hoc subgroup analysis of the April 2004; published online 1 July 2004 GREACE results, we report the long-term effect of Statins and ACE inhibitors VG Athyros et al 782 combined treatment with a statin plus an ACEI in Endpoint adjudication comparison to each drug alone or neither drug, There was an independent adjudication committee regardless of the initial assignment of patients in the blinded to treatment assignment. All end points structured or usual care groups. were hard end points and required hospitalisation. Total and cardiac mortality, and had to be diagnosed by a cardiology Study population—methods clinic to be considered as end points. Death Original study certificates, discharge summaries, ECG, CT Scan or MRI were used to validate an end point. Patients Study design and patients with an end point were considered only once; even Recruitment of patients started 6 years ago and was if they had several CHD events. completed within a 2-year period.6–9 Only patients with established CHD were included: history of Protocol prior myocardial infarction or 470% stenosis of at All patients with a LDL-C 42.6 mmol/l (100 mg/dl) least one coronary artery, as documented by a were enrolled into the study. In the atorvastatin coronary angiogram. Patients with recent acute group, the starting dose was 10 mg/day. If the NCEP coronary syndromes were not excluded. Inclusion LDL-C goal of o2.6 mmol/l was not reached within criteria were age o75 years, LDL-C 42.6 mmol/l 6 weeks, the dose of atorvastatin was increased to (100 mg/dl) and (TG) o4.5 mmol/l 20 mg/day. With evaluations every 6 weeks the dose (400 mg/dl). Exclusion criteria were renal or of atorvastatin was titrated up to 80 mg/day for dysfunction, prior hypolipidaemic treatment, child- patients not reaching the LDL-C goal with lower bearing potential and any significant disease likely dosages. The patients on usual care received to limit life to less than the duration of the study, whatever drug treatment was prescribed by their such as and heart failure New York physician. Heart Association class III or IV. Patients that were The study was powered based on a comparison of scheduled for coronary revascularization were also the estimated proportion of patients that would excluded. All consecutive patients referred to our experience a cardiovascular event in the atorvastatin out-patient clinic were enrolled if eligible. Some of and usual care groups. Data from a large-scale them (12%) had an MI (1–3 years previously) and epidemiologic study involving an adult Greek some (8%) were included 7–10 days after admission population (unpublished) provided the estimate of unstable . Most of the patients were that Greek CHD patients under usual care have an recently diagnosed. If the patients diagnosed as all-cause event rate of 24% over a 3-year period. We having CHD more than 1 year before enrolment were considered that our findings would be significant if excluded, the mean time from diagnosis to enrol- treatment with atorvastatin to NCEP LDL-C goal ment was 26 days. All patients attended the produced a relative 30% reduction in all cause Hipocration Hospital. Equal numbers of patients events. Using this hypothesis, we performed the were then randomly allocated to atorvastatin treat- two-sample z-test for comparing two binomial ment based in our outpatient clinic or to usual care parameters. This suggested that 800 patients per outside the hospital. Randomisation of patients was group would provide a 90% power, with a type I carried out in cooperation with the Greek Society of error rate of 5%. No reduction in sample size due to General Practitioners, which also carried out the patients not completing the study for personal or follow-up of the usual care patients, while the medical reasons was calculated because this was an University Clinic was responsible for the atorvasta- intention-to-treat study. tin-treated patients. Cardiologists or general practi- GREACE demonstrated significant reductions in tioners of the patients’ choice treated those on usual morbidity and mortality associated with structured care according to their own standards of secondary management of dyslipidaemia with dose titration of CHD prevention. There were no limitations in the atorvastatin (10–80 mg/day, mean dose 24 mg/day). treatment of usual care patients. This could include In the structured care arm, 98% of patients were on life style changes, such as hypolipidaemic , long-term treatment with atorvastatin and 95% weight loss, plus all necessary drug treat- reached the NCEP LDL-C treatment target ment, including -lowering agents. Atorvastatin o2.6 mmol/l; 100 mg/dl). In the usual care arm, was not excluded from the usual care group. One of only 12% of patients were on statins and 3% the objectives of the study was to estimate the reached the NCEP LDL-C treatment target. difference in lipid lowering treatment within two settings: a specialist unit with a strict protocol dictating to treat to the NCEP LDL-C target and usual Post-hoc analysis care outside the hospital. In this way, we tried to point out the benefit of reaching this specific In the present post hoc subgroup analysis, we treatment target in CHD patients. The study received assessed the effect of the combination of a statin ethical approval and informed consent was obtained plus an ACEI, in comparison to each drug alone or from all patients before enrolment. neither drug in four groups of patients, regardless of

Journal of Human Hypertension Statins and ACE inhibitors VG Athyros et al 783 their assignment to structured or usual care in the assessed at baseline, at the 6th treatment week and original study. Group A included patients on a statin every 6 months thereafter. Biochemical measure- and ACEI, group B included patients on a statin but ments were made on each serum sample using an not an ACEI, group C patients were taking an ACEI Olympus AU 560 autoanalyser and appropriate but not a statin, and group D patients were neither reagents (Olympus GmbH, Clare, Ireland). LDL-C on a statin or an ACEI. The demographic character- was calculated using the Friedewald formula (ap- istics and baseline CHD risk factors of the four plicable in all patients, since they had serum groups are shown in Table 1 and the baseline levels o4.5 mmol/l (400 mg/dl). SCr lipid values in Table 2. The study population had was measured using the Jaffe method (reference a high risk of events because they all had CHD. range: 53–115 mmol/l; 0.6–1.3 mg/dl). CrCl was esti- Furthermore, the baseline LDL-C level (4.6 mmol/l; mated from SCr using the Cockroft–Gault formula, 178 mg/dl) was elevated, 20% of patients had which corrects for age, weight and gender.11 SUA mellitus, 43% were hypertensive and was assessed with an enzymatic colorimetric 35% had a revascularization prior to entry into the test (uricase); reference range: 150–415 mmol/l study. (2.5–7.0 mg/dl). All the patients participating in Lipoproteins, serum creatinine (SCr), creatinine the original study were included in the present (CrCl), and serum (SUA) were post hoc subanalysis.

Table 1 Baseline characteristics and coronary heart disease risk factors in the four treatment groups

Group A Group B Group C Group D P-value n ¼ 460 n ¼ 420 n ¼ 371 n ¼ 349

Men/women 79/21% 78/22% 77/23% 80/20% NS Age (years) 60712 58711 59712 58714 NS Body mass index (kg/m2)2571.0 2471.1 2470.9 2571.1 NS Myocardial infarction 82% 81% 82% 80% NS Diabetes mellitus 21% 20% 19% 20% NS Arterial hypertension 43% 41% 45% 43% NS PTCA/CABG 34% 36% 36% 35% NS Recent unstable angina 8% 9% 8% 7% NS

Age and body mass index are expressed as mean values 7 1 s.d. and the rest as percentage of patients. PTCA ¼ percutaneous transluminal coronary angioplasty, CABG ¼ coronary artery bypass grafting.

Table 2 : low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglycerides (TG), systolic blood pressure (SBP), diastolic blood pressure (DBP), creatinine clearance (CrCl), and serum uric acid (SUA) (mean values 7 1 s.d.) of the four treatment groups at baseline and during treatment

Group A Group B Group C Group D ANOVA n ¼ 460 n ¼ 420 n ¼ 371 n ¼ 349

Baseline LDL-C (mmol/l) 4.7670.80 4.6570.75 4.6870.72 4.7170.85 NS On-study LDL-C (mmol/l) 2.5970.21 2.5670.26 4.4570.65 4.5070.67 o0.0001 P-value o0.0001 o0.0001 NS NS Baseline HDL-C (mmol/l) 0.9870.16 0.9670.18 0.9670.13 0.9870.18 NS On-study HDL-C (mmol/l) 1.0970.18 1.0670.16 0.9670.13 1.0170.16 ¼ 0.02 P-value ¼ 0.002 ¼ 0.001 NS NS Baseline TG (mmol/l) 2.0870.72 2.0570.80 2.0770.73 2.0370.71 NS On-study TG (mmol/l) 1.3970.33 1.4570.35 1.9670.75 1.9370.72 o0.0001 P-value o0.0001 o0.0001 NS NS Baseline SBP (mmHg) 124717 123714 123716 125716 NS On-study SBP (mmHg) 120713 122712 121711 122713 NS P-value NS NS NS NS Baseline DBP (mmHg) 7679747875797579NS On-study DBP (mmHg) 7577747874777378NS P-value NS NS NS NS Baseline CrCl (ml/min) 76713 75711 77712 76713 NS On-study CrCl (ml/min) 8478827874711 73711 o0.0001 P-value o0.0001 o0.0001 0.008 0.01 Baseline SUA (mmol/l) 428759 434759 422759 416759 NS On-study SUA (mmol/l) 393754 387748 440765 434759 o0.0001 P-value o0.0001 o0.0001 0.02 0.01

To convert data from mmol/l to mg/dl multiply LDL and HDL cholesterol values by 38.7 and TG by 88.6. For SUA to convert data from mmol/l to mg/dl divide by 59.48.

Journal of Human Hypertension Statins and ACE inhibitors VG Athyros et al 784 Table 3 Medical therapy for at least 6 months at any point during the study or until the time of their death in cases of patients that died before the 6th treatment month in the four treatment groups

Group A Group B Group C Group D P-values n ¼ 460 n ¼ 420 n ¼ 371 n ¼ 349

Aspirin or other antiplatelet agents 89% 90% 87% 86% NS Beta-blockers 85% 87% 85% 84% NS 14% 15% 15% 16% NS Calcium channel blockers 26% 25% 27% 28% NS 12% 11% 12% 13% NS Hypolipidaemic 100% 100% 0% 0% o0.0001 ACE inhibitors 100% 0% 100% 0% o0.0001

ACE ¼ angiotensin-converting enzyme.

Medical therapy From those on statins, 807 were on atorvastatin The long-term medical treatment (drug treatment for (783 from the structured care group, mean dose at least 6 months at any point of the study) is shown 24 mg/day and 24 from the usual care group, mean in Table 3. For patients who died before the dose 15 mg/day). The remaining patients (n ¼ 73) completion of 6 months of the study, their drug were on various statins: 40 were on , treatment until the time of their death was recorded. mean dose 20 mg/day, 25 on , mean dose This procedure was also used for patients that died 24 mg/day and eight were on , mean dose before the completion of the 3-year follow-up. No 40 mg/day (all in the usual care group). patient was excluded from the final analysis. From those on an ACEI, 328 (40%) were on enalapril (mean dose 11 mg/day), 293 (35%) were on End points quinapril (mean dose 12 mg/day), and 210 (25%) In this post hoc analysis of the GREACE Study, we were on perindopril (4 mg/day). The GREACE study used the composite end point ‘all events’ compris- was initiated 6 years ago and that is why newer ing all primary end points of the original study: all- ACEI and angiotensin blockers were not cause and coronary mortality, coronary morbidity prescribed. (nonfatal myocardial infarction, revascularization, From the 880 patients on statins, 460 were on unstable angina and congestive heart failure) and various ACEIs (Group A) and 420 were not (Group stroke. No end point included in the original study B). From the 720 patients that were not on statins, was left out and no new end point was included. 371 were on various ACEIs (Group C) and 349 were not (Group D). Statistical analyses Analysis of variance (ANOVA) was used to assess differences in lipid, SCr, CrCl and SUA values within and between treatment groups as well as Lipid values primary end point rates between the four treatment Lipid values at baseline and on-study as well as groups. The Mantel–Haenszel method was used to their changes are shown in Table 2. As expected, calculate relative risk reductions (RRR) between Groups A and B had significant changes in lipid treatment groups. A two-tailed value of Po0.05 was values, especially in LDL-C (Table 2). In contrast, considered statistically significant. The Statgraphics patients from Groups C and D who were not on Plus (Statgraphics, Rockville, MD, USA) program statins did not show significant changes in lipid was used for all statistical analyses. values during study (Table 2).

Blood pressure Results There were no significant differences in systolic or There were no significant differences between the diastolic blood pressure between the four treatment four treatment groups in demographic characteris- groups either at baseline or during the study tics and in baseline CHD risk factors (Table 1) or (Table 2). biochemical parameters (Table 2). There were also no significant differences in drug treatment during Renal function the study, except for the use of statins or ACEIs There were no significant differences in baseline (Table 3). CrCl between the four treatment groups (Table 2). From the entire study population (n ¼ 1600), 880 During the study, CrCl in Groups A and B (on patients (55%) were on statins and 720 (45%) were statins) was significantly increased (Table 2), while not, while 831 (52%) were on ACEI and 769 (48%) it was reduced in Groups C and D (not on statins) were not. (Table 2). There were no significant differences in

Journal of Human Hypertension Statins and ACE inhibitors VG Athyros et al 785 Table 4 Comparisons of the percent relative risk reductions (RRR) and 95% confidence intervals (CI) for the primary end point and its individual components in the four treatment groups

Cardiovascular death + nonfatal MI Cardiovascular death + nonfatal MI + revascularization All events

RRR (95% CI) RRR (95% CI) RRR (95% CI) P-value P-value P-value

Group A vs B À34% (À62–14%) À38% (À61–À3%) À31 % (À48–À6%) 0.08 0.02 0.01 Group A vs C À63% (À78–À38%) À63% (À76–À44%) À59% (À72–À48%) o0.0001 po0.0001 o0.0001 Group A vs D À71% (À82–À54%) 70% (À82–À57%) À63% (À74–À51%) o0.0001 po0.0001 o0.0001 Group B vs C À44% (À65–À10%) À40% (À59–À13%) À42% (À48–À19%) 0.01 0.004 0.0001 Group B vs D À57% (À73–À32%) À55% (À68–À35%) À50% (À65–À36%) o0.0001 o0.0001 o0.0001 Group C vs D À23% (À48–12%) À22% (À38–6%) À9% (À27–10%) 0.1 0.06 0.1

MI ¼ myocardial infarction, ACEI ¼ angiotensin-converting enzyme inhibitor, Group A ¼ Statin + ACEI, Group B ¼ Statin no ACEI, Group C ¼ ACEI no statins, Group D ¼ no ACEI no statins.

SUA levels at baseline between the four treatment comparison to Group C (no statin, on ACEI), and groups (Table 2). However, the mean on treatment 5.4. in comparison to Group D (no statin, no ACEI). values were significantly different: a reduction in In Group B, 7.4 patients needed to be treated with a Groups A and B and increase in Groups C and D statin to avoid one cardiovascular event, in compar- (Table 2). ison to Group C, and 10.8 patients in comparison to Group D. End points During the 3-year follow-up there were 292 cardio- Other factors that may have influenced vascular events (primary end point). There were 45 clinical outcomes (10% of patients) in group A, 61 (14.5%) in group B, In this open label study compliance to all drugs was 91 in Group C (24.5%), and 95 events in the Group D evaluated by recording the frequency of prescribing (27% of patients). Relative risk reductions (RRR-and in a personal health book. We report on-study 95% confidence interval (CI) in the primary end treatment in Table 3 according to this information. point between Group A and all the rest was There were no significant differences between the significant (Table 4). The RRR in Group A (statin four treatment groups in demographic characteris- plus ACEI) vs group C (no statin, on ACEI) was tics and CHD factors at baseline (Table 1), in À59% (95% CI À72 to À48%, Po0.0001) and baseline lipid values and renal function (Table 2), À63% (95% CI À74 to À51%, Po0.0001) vs group in concomitant drug treatment during the study D (no statin, no ACEI). The RRR in Group A (statin (Table 3), and level of glycemic control6–9 or in blood plus ACEI) vs Group B (statin, no ACEI) was À31% pressure during the study (Table 2). Both at entry (95% CI À48 to À6%, P ¼ 0.01). The RRR in Group B and during the study, smokers were similarly vs Group C was 42% (95% CI À48 to À19%, distributed in the four treatment groups.6–9 All P ¼ 0.0001) and vs Group D it was 50% (95% CI patients in the four treatment groups received À65 to À36%, Po0.0001). There was no significant advice on life-style changes and the body mass difference in RRR between Groups C and D (À9%, index (an approximate index of compliance) at 95% CI À27 À10%, P ¼ 0.1) (Table 4). In order to baseline and during the study was similar in all provide an idea of the contribution of the compo- groups. Thus, the beneficial effect on clinical nents of the primary end point to event rate outcomes should mainly be attributed to statin or reduction, we also provide (Table 4) the percent ACEI treatment or both. RRRs (and 95% CI) for cardiovascular death þ nonfatal MI, cardiovascular death þ nonfatal MI þ revascularization (widely used composite pri- Discussion mary end points in other studies). In the present analysis involving high-risk dyslipi- Numbers needed to treat (NNT) daemic CHD patients, treatment with a statin plus In Group A, 17 patients needed to be treated with a an ACEI significantly reduced cardiovascular events statin þ ACEI combination for a period of 3 years to more than each drug alone or neither drug (Table 4). avoid one cardiovascular event, in comparison with The RRR of the primary end point in patients on Group B (on statin, no ACEI), 7.1 patients in statin treatment (Groups A and B) in comparison to

Journal of Human Hypertension Statins and ACE inhibitors VG Athyros et al 786 those not treated for their dyslipidaemia (Groups C Various mechanisms20,21 have been proposed to and D) was expected. However, the RRR in the explain the decrease in ischemic events seen with primary end point in patients on the statin þ ACEI ACEIs in heart failure patients with or without combination (Group A) in comparison to patients on CHD22,23 or those with preserved left ventricular a statin alone (Group B) also showed a significant function.24 The Trandolapril Cardiac Evaluation benefit (Table 4), suggesting a synergistic effect of (TRACE) Study25 reported that long-term treatment the two drug classes. This has not been previously with trandolapril in patients with reduced left shown in an end point study. ventricular function soon after MI significantly In contrast, treatment with an ACEI but no statin reduced the risk of overall mortality, mortality from (Group C) compared with those not taking either of cardiovascular causes, sudden death and the devel- these drugs (Group D) did not produce a significant opment of severe heart failure. The Trial on Rever- benefit in clinical outcome (Table 4). The EUROPA sing (TREND)26 showed Trial5 showed a substantial benefit with a high dose that the ACEI quinapril improved endothelial of perindopril (8 mg/day) in 6110 patients vs 6108 dysfunction in patients who were normotensive patients on (a total of 12218 patients in two and who did not have severe hyperlipidaemia or groups). Our analysis included 1600 patients and evidence of heart failure. The Heart Outcomes only half of them (about 400 patients on ACEI þ Prevention Evaluation (HOPE) study24 confirmed statin and another 400 on ACEI alone) were on low the benefits of ramipril in patients aged 55 years or doses of various ACEIs. Despite these small numbers older, with normal left ventricular function, at high there was an event rate reduction when comparing risk of cardiovascular complications (high preva- the ACEI alone vs the no ACEI þ no Statin group. lence of diabetes, hypertension, stroke, and obstruc- This difference (eg À24% in cardiovascular death þ tive peripheral vascular disease). In this high-risk nonfatal MI) may have been statistically significant cohort, ramipril reduced the risk of MI, worsening if the study had included more patients. As expected, and new angina, and the occurrence of coronary the Groups on statins (A and B) had significant revascularization. In the HOPE Trial, only 28% of reductions in LDL-C levels, a major modulator of patients were on lipid lowering treatment. The clinical benefit in CHD patients, while ACEI use had Perindopril Protection Against Recurrent Stroke a neutral effect on lipid values (Table 2). Study (PROGRESS)27 provided definitive evidence In statin-treated patients there was a significant that blood pressure lowering in patients with improvement in renal function and a reduction in previous stroke or transient ischaemic attack sig- SUA levels (Table 2), which are considered to be nificantly reduces the incidence of secondary stroke CHD risk factors, even within the upper limits of the and major vascular events. There was no report on reference range.12–14 There are now several reports how many patients were on statins in the PRO- on the effects of statins (mainly atorvastatin and GRESS trial. The EUROPA Trial showed that 8 mg/ simvastatin) on renal function and SUA levels as day of perindopril can significantly improve out- well as their contribution to CHD risk reduc- come in patients with stable CHD without apparent tion.8,9,15–19 ACEIs (Group A and C) had a neutral heart failure.5 The benefits reported for perindopril effect on renal function and SUA levels (Table 2). were in addition to other preventive measures, These differences in renal function may have including , beta-blockers, and lipid-lowering influenced the RRRs. drugs, and were consistent for all patients. Interest- Clinical benefit could not be attributed to a ingly, 58% of patients in the EUROPA Trial patients reduction in blood pressure because ACEI-treated were on statins, in both treatment groups. Therefore, patients (Groups A and C) had a similar systolic and it would be useful to carry out a subgroup analysis of diastolic blood pressure as the non-ACEI-treated EUROPA. The benefit of the statin þ ACEI combina- patients (Groups B and D). This could be attributed tion was assessed in the Simvastatin/Enalapril to several reasons. ACEI were administered at low Coronary Atherosclerosis Trial (SCAT).28 This was doses and patients in the non-ACEI groups were on an angiographic study (2 Â 2 factorial design). The higher doses of beta-blockers, CCBs or diuretics. main finding was that lipid-lowering therapy for 3–5 ACEIs were prescribed not only for hypertension but years with simvastatin resulted in significant slow- also for secondary CHD prevention, mostly in ing of coronary atherosclerosis in normolipidaemic patients with an anterior wall MI (this is the main CHD patients. Enalapril (an ACEI) had a neutral reason for the low doses of ACEI) and after an MI effect on CHD. Adding enalapril to simvastatin did (especially of the anterior wall) blood pressure is not result in an incremental angiographic benefit. usually lower than before the MI. We also did not However, coronary angiography has its limitations29 find any additive effect of statins in reducing blood and the participants were ‘low risk’ CHD patients. pressure (Group A). This may be because all patients Furthermore, SCAT was not an end point study and were on treatment for secondary CHD prevention it only included 460 patients. In contrast to SCAT, with several antihypertensives, thus blunting any another study showed a synergistic effect of simvas- hypotensive effect of statins. Other risk factors that tatin and enalapril in increasing the response to could influence outcome were similar in all four postischaemic vasodilatation in hypercholesterolae- groups (Tables 1 and 3). mic patients.30

Journal of Human Hypertension Statins and ACE inhibitors VG Athyros et al 787 It is likely that the synergistic/additive effect of coronary events after myocardial infarction in patients statins and ACEI we found could be attributed to the with average cholesterol levels. N Engl J Med 1996; complementary pleiotropic effects of these 335: 1001–1009. drugs.31,32 For example, ACEIs and statins can have 3 Long-term Intervention with Pravastatin Ischaemic beneficial effects on haemostatic and other fac- Disease (LIPID) Study Group. Prevention of cardiovas- cular events and death with pravastatin in patients tors.33–36 Moreover, it has been shown that treatment 37,38 39,40 with coronary heart disease and a broad range of initial with ACEIs and statins improved vasodilata- cholesterol levels. N Engl J Med 1998; 339: 1349–1357. tion and decreased vascular damage. A significant 4 Heart Protection Study Collaborative Group. MRC/BHF additive effect on structural vascular damage, blood Heart Protection Study of cholesterol lowering with pressure, and vascular resistance was also shown simvastatin in 20536 high-risk individuals: a rando- during combination treatment of simvastatin and mized placebo-controlled trial. Lancet 2002; 360: enalapril in hypertensive subjects.30,32 We reported 7–22. that quinapril41 and atorvastatin42,43 have a benefi- 5 Fox KM, and The EURopean trial On reduction of cial effect on heart rate variability and aortic cardiac events with Perindopril in stable coronary elasticity, left ventricular ejection fraction and left Artery disease Investigators. of perindopril in reduction of cardiovascular events among patients ventricular mass index.43 All these factors could 44–48 with stable : randomised, account for an increased event rate. double-blind, placebo-controlled, multicentre trial (the EUROPA study). Lancet 2003; 362: 782–788. 6 Athyros VG et al. Treatment with atorvastatin to the Study limitations National Cholesterol Educational Program goals versus usual care in secondary Coronary Heart Disease This post hoc analysis was not double-blind and prevention: the GREek Atorvastatin and Coronary- placebo-controlled, because of ethical and practical heart-disease Evaluation (GREACE) Study. Curr Med restrictions. The main goal was to assess the clinical Res Opin 2002; 18: 220–228. benefit from NCEP guideline implementation in 7 Athyros VG et al. Early benefit from structured care comparison to that seen with real-life treatment with atorvastatin in patients with coronary heart patterns. GREACE was an unsponsored target-based disease and diabetes mellitus: a subgroup analysis of study. the GREek Atorvastatin and Coronary-heart-disease Evaluation (GREACE) Study. Angiology 2003; 54: 679–690. Conclusions 8 Athyros VG et al. The effect of statins versus untreated dyslipidaemia on renal function in patients with In dyslipidaemic CHD patients the combination of a coronary heart disease: a subgroup analysis of the statin plus an ACEI reduces cardiovascular events GREek Atorvastatin and Coronary-heart-disease Eva- more than a statin alone and substantially more than luation (GREACE) Study. J Clin Pathol 2004 (in press). an ACEI alone. Such a combination should be 9 Athyros VG et al. The effect of statins versus untreated dyslipidaemia on serum uric acid levels in patients considered when treating high-risk CHD patients. with coronary heart disease: a subgroup analysis of the Treatment with an ACEI but not a statin in GREek Atorvastatin and Coronary-heart-disease Eva- comparison to patients not treated with either drug luation (GREACE) Study. Am J Kidney Dis 2004; 43: did not significantly reduce clinical events, at least 589–599. in these small groups of patients. These relation- 10 Executive Summary of The Third Report of The ships need to be assessed in the trials already National Cholesterol Education Program (NCEP). Ex- published and should be considered when design- pert Panel on detection, evaluation, and treatment of ing future trials. high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001; 285: 2486–2497. 11 Cockroft DW, Gault MH. Prediction of creatinine Declaration of interest clearance from serum creatinine. Nephron 1976; 16: 31–41. The GREACE study was conducted independently; 12 Kaplan RC, Heckbert SR, Furberg CD, Psaty BM. no Company or Institution has supported it finan- Predictors of subsequent coronary events, stroke, and cially. Some of the authors have attended confer- death among survivors of first hospitalized myocardial infarction. J Clin Epidemiol 2002; 55: 654–664. ences and participated in other trials sponsored by 13 Bickel C et al. Serum uric acid as an independent various pharmaceutical companies. predictor of mortality in patients with angiographi- cally proven coronary artery disease. Am J Cardiol 2002; 89: 12–17. References 14 Fang J, Alderman MH. Serum uric acid and cardiovas- cular mortality the NHANES I epidemiologic follow- 1 Scandinavian Simvastatin Survival Study (4S) Group. up study, 1971–1992. National Health and Nutrition Randomized trial of cholesterol lowering in 4444 Examination Survey. JAMA 2000; 283: 2404–2410. patients with coronary heart disease. Lancet 1994; 15 Elisaf M, Mikhailidis DP. Statins and renal function. 344: 1383–1389. Angiology 2002; 53: 493–502. 2 Sacks FM, et al, for the Cholesterol and Recurrent 16 Youssef F et al. The early effect of lipid-lowering Events Trial Investigators. The effect of pravastatin on treatment on carotid and femoral intima media thick-

Journal of Human Hypertension Statins and ACE inhibitors VG Athyros et al 788 ness (IMT). Eur J Vasc Endovasc Surg 2002; 23: 32 Nazzaro P et al. Distinct and combined vascular effects 358–364. of ACE blockade and HMG-CoA inhibition 17 Youssef F et al. The effect of short-term treatment with in hypertensive subjects. Hypertension 1999; 33: simvastatin on renal function in patients with periph- 719–725. eral arterial disease. Angiology 2004; 55: 53–62. 33 Papadakis JA et al. Effect of hypertension, and its 18 Sinzinger H, Kritz H, Furberg CD. Atorvastatin reduces treatment, on lipid, lipoprotein (a), fibrinogen and microalbuminuria in patients with familial hypercho- bilirubin levels in patients referred for dyslipidaemia. lesterolemia and normal tolerance. Med Sci Am J Hypertens 1999; 12: 673–681. Monit 2003; 9: 88–92. 34 Milionis HJ, Elisaf MS, Mikhailidis DP. The effects of 19 Heart Protection Study Collaborative Group. MRC/ lipid-regulating therapy on haemostatic parameters. BHF Heart Protection Study of cholesterol-lowering Curr Pharm Des 2003; 9: 2425–2443. with simvastatin in 5963 people with diabetes: a 35 Tsiara S, Elisaf M, Mikhailidis DP. Early vascular randomised placebo-controlled trial. Lancet 2003; benefits of statin therapy. Curr Med Res Opin 2003; 19: 361: 2005–2016. 540–556. 20 Lonn EM et al. Emerging role of angiotensin-convert- 36 Ganotakis ES, Papadakis JA, Vrentzos GE, Mikhailidis ing enzyme inhibitors in cardiac and vascular protec- DP. The effects of antihypertensive therapy on haemo- tion. Circulation 1994; 90: 2056–2069. static parameters. Curr Pharm Des 2003; 9: 2445–2464. 21 Dzau VJ. Mechanism of protective effects of ACE 37 Virdis A et al. Presence of cardiovascular structural inhibition on coronary artery disease. Eur Heart J changes in essential hypertensive patients with cor- 1998; 19(Suppl J): J2–J6. onary microvascular disease and effects of long-term 22 Yusuf S et al. Effect of enalapril on myocardial therapy. Am J Hypertens 1996; 9: 361–369. infarction and unstable angina in patients with low 38 Novo S, Abrignani MG, Corda M, Strano A. Cardio- ejection fractions. Lancet 1992; 340: 1173–1178. vascular structural changes in hypertension: possible 23 Pfeffer MA et al; The SAVE Investigators. Effect of regression during long-term antihypertensive treat- on mortality and morbidity in patients with ment. Eur Heart J 1991; 12(Suppl G): 47G–53G. left ventricular dysfunction after myocardial infarc- 39 Eichstadt HW et al. Improvement of myocardial tion. N Engl J Med 1992; 327: 669–677. perfusion by short-term fluvastatin therapy in coronary 24 The Heart Outcomes Prevention Evaluation Study artery disease. Am J Cardiol 1995; 76(Suppl A): (HOPE) Investigators. Effects of an angiotensin-con- 122A–125A. verting enzyme inhibitor, ramipril, on cardiovascular 40 Cheng KS, Mikhailidis DP, Hamilton G, Seifalian AM. events in high-risk patients. N Engl J Med 2000; 342: A review of the carotid and femoral intima-media 145–153. thickness as an indicator of the presence of peripheral 25 Kober L et al. A of the angiotensin- vascular disease and cardiovascular risk factors. converting-enzyme inhibitor trandolapril in patients Cardiovasc Res 2002; 54: 528–538. with left ventricular dysfunction after myocardial 41 Kontopoulos AG, Athyros VG, Papageorgiou AA, infarction. Trandolapril Cardiac Evaluation (TRACE) Boudoulas H. Effect of quinapril or metoprolol on Study Group. N Engl J Med 1995; 333: 1670–1676. circadian sympathetic and parasympathetic modula- 26 Mancini GB et al. Angiotensin-converting enzyme tion after acute myocardial infarction. Am J Cardiol inhibition with quinapril improves endothelial vaso- 1999; 84: 1164–1169. motor dysfunction in patients with coronary artery 42 Pehlivanidis AN et al. Heart rate variability after long- disease. The TREND (Trial on Reversing ENdothelial term treatment with atorvastatin in hypercholestero- Dysfunction) Study. Circulation 1996; 94: 258–265. laemic patients with or without coronary artery 27 PROGRESS Collaborative Group. Randomised trial of a disease. Atherosclerosis 2001; 157: 463–469. perindopril-based blood-pressure-lowering regimen 43 Kontopoulos AG et al. Long-term treatment effect of among 6,105 individuals with previous stroke or atorvastatin on aortic stiffness in hypercholesterolae- transient ischaemic attack. Lancet 2001; 358: mic patients. Curr Med Res Opin 2003; 19: 22–27. 1033–1041. 44 Mitchell LB et al; AVID Investigators. Are lipid- 28 Teo KK et al. Long-term effects of cholesterol lowering lowering drugs also antiarrhythmic drugs? An and angiotensin-converting enzyme inhibition on analysis of the Antiarrhythmics versus Implantable coronary atherosclerosis: the Simvastatin/Enalapril Defibrillators (AVID) trial. J Am Coll Cardiol 2003; 42: Coronary Atherosclerosis Trial (SCAT). Circulation 81–87. 2000; 102: 1748–1754. 45 Quintana M et al. Heart rate variability as a means of 29 Schoenhagen P, White RD, Nissen SE, Tuzcu EM. assessing prognosis after acute myocardial infarction. Coronary imaging: angiography shows the stenosis, but Eur Heart J 1997; 18: 789–797. IVUS, CT, and MRI show the plaque. Cleve Clin J Med 46 van Boven AJ et al. Depressed heart rate variability is 2003; 70: 713–719. associated with events in patients with stable coronary 30 Esper RJ et al. -dependent responses in artery disease and preserved left ventricular function. patients with hypercholesterolemic coronary artery REGRESS Study Group. Am Heart J 1998; 135: disease under the effects of simvastatin and enalapril, 571–576. either separately or combined. Am Heart J 2000; 140: 47 Stefanadis C et al. Aortic stiffness as a risk factor for 684–689. recurrent acute coronary events in patients with 31 Gryglewski RJ et al. Comparison of endothelial ischaemic heart disease. Eur Heart J 2000; 21: 390–396. pleiotropic actions of angiotensin converting enzyme 48 Benetos A et al. Pulse pressure: a predictor of long- inhibitors and statins. Ann N Y Acad Sci 2001; 947: term cardiovascular mortality in a French male 229–245. population. Hypertension 1997; 30: 1410–1415.

Journal of Human Hypertension