Renin-angiotensin system blockade: to what extent?

Abstract

Two experimental studies reported in this issue of the journal [1,2] describe different methods for blocking the renin–angiotensin system (RAS) and analyse the cardiac and renal consequences of these methods. Both reports consider the most effective way to inhibit the RAS either by an angiotensin-converting enzyme (ACE) inhibitor, an angiotensin II receptor (AT1R) antagonist, or their combination. These two studies have similar methodological strengths: a wide range of studied doses, a sufficient number of animals per group, and the selection of robust end-points. The synergistic effects on blood pressure and left ventricular mass (LVM) of treatment combining an ACE inhibitor and AT1R antagonist at various doses are precisely documented by the two studies. They reach similar conclusions concerning the candesartan and ramipril combination [1], and the losartan and enalapril combination [2]. The German study makes two specific contributions. (i) A demonstration that the synergistic effects on blood pressure and LVM reduction observed are due to the combination of an ACE inhibitor with an AT1R antagonist, blocking the RAS at two successive sites along the renin-dependent angiotensin II-producing pathway. No such synergism is observed by simply combining a RAS blocker with an antihypertensive drug, as the T-calcium-channel blocker mibefradil, for example, does not enhance the effects of candesartan in the way that ramipril does. (ii) When high doses of each single site RAS blocker are combined, the synergistic and, even more, the additive effects of the combination are no longer observed. This was already demonstrated by Morgan et al. [3] in acute experiments. The French study investigated the influence of salt intake on blood pressure and LVM responses to various doses of losartan, enalapril and their combination in spontaneously hypertensive rats (SHR). Doses of an ACE inhibitor and an AT1R antagonist in a combination that was one-third to one-tenth of those used in a single-drug treatment were found to have the same haemodynamic and trophic effects as the highest doses of a single drug (100 mg/kg). Independently of the site of RAS blockade, all haemodynamic changes are qualitatively similar and proportional to the intensity of the blockade. This study documents the renal risk of treatment with a 10 mg/kg enalapril–losartan combination in SHR with low salt intake (0.05% Na+). Such a risk has already been reported in SHR with normal salt intake (0.3%) [4]. Finally, the results indicate that, on a high sodium intake (2%), haemodynamic effects similar to those observed a low sodium intake are observed, but the doses of the RAS blockers required are higher. Above a certain dose, the effect appears to be maximal, and reaches a plateau. The 10 mg/kg enalapril–losartan combination, poorly tolerated on a low-salt diet, is well tolerated on a high sodium intake and is almost as effective as the 30 mg/kg combination. By contrast, no plateau effect is found when vascular reactivity to exogenous angiotensin I and II is analysed. However, we still ignore whether and to what extent this pharmacological response is related with the blood pressure-independent beneficial effect of neutralization of the direct tissue effects of angiotensin II on target organs. This observation could be interpreted, with caution, as evidence that it is possible to obtain supplementary beneficial effects by reinforcing the RAS blockade, even in the absence of a supplementary lowering effect on blood pressure [5]. The theoretical and practical implications of these two experimental papers must be linked to the results of a series of previous clinical investigations [6,7]. At the time when an ACE inhibitor dissociates from ACE active sites and angiotensin II reappears in the presence of an increased level of plasma and interstitial angiotensin I due to the interruption of the angiotensin II-renin feedback, the concurrent administration of an AT1R antagonist protects the AT1R from newly produced agonist. Reciprocally, when less AT1R antagonist is bound to the AT1Rs, an ACE inhibitor reduces the production of angiotensin II available to compete with the antagonist. The demonstration of this physiological mechanism, intrinsic to the RAS, provides a more likely explanation for the reinforcement of the RAS inhibition by a dual blockade than the putative intervention of angiotensin II production by ACE-independent pathways [8]. This angiotensin II-renin feed-back counterregulatory mechanism is more important when the dosing interval is 24 h (once daily prescriptions) or more (non-compliance) than when it is short (b.i.d. or t.i.d. prescriptions) [9,10]. Supplementary beneficial effects of a combined blockade on blood pressure and proteinuria have been observed in hypertensive and diabetic patients [11,12]. Convincing results have now been obtained in several large-scale randomized trials, which have demonstrated that an intensive RAS inhibition, including dual blockade, prevents cardiovascular and renal damage. Interest in the reinforcement of the RAS blockade first showed promise in studies in animal models that demonstrated a dose-dependent relationship between ACE inhibition and the regression of vascular [13,14], cardiac [15] and renal lesions [16]. With the notable exception of the beneficial effect on left ventricular mass reported at low daily doses of ramipril in one model of aortic banding in rats [17], all studies were in favour of using high doses of RAS blockers to maximize the benefits in terms of preventing damage to target organs. This need for high doses of a single-site RAS blocker to prevent target organ damage highlights the potential value of combining much lower doses of two different RAS blockers, which have consistently been shown to neutralize the RAS more effectively than higher doses of a single-site RAS blocker [18–21]. The results of several large-scale secondary prevention trials have now convincingly demonstrated the importance of dose [10]: high doses of ACE inhibitors, such as ramipril and perindopril, and of AT1R antagonists, such as irbesartan, are more effective than low doses. In the PROGRESS trial, a 4 mg daily dose of perindopril was not effective on the combined primary end-point in hypertensive and non-hypertensive patients with a history of stroke or transient ischaemic attack [22]. By contrast, in the EUROPA trial, a 8 mg daily dose of perindopril effectively reduced the incidence of the combined primary end-point in patients with stable coronary heart disease [23]. In type 2 diabetic patients recruited in the MICROHOPE study [24], a 10 mg daily dose of ramipril was effective on the combined primary end-point. A 1.25 mg daily dose of ramipril was not effective on the combined end-point in the DIABHYCAR study [25], nor was a 2.5 mg daily dose in the SECURE trial [26], a substudy of the HOPE trial on carotid artery atherosclerosis progression [27]. As for AT1R antagonists, irbesartan was more effective at a dose of 300 mg per day than at a dose of 150 mg per day in the IRMA II study for preventing overt diabetic nephropathy in microalbuminuric patients with type 2 diabetes [28]. In patients with chronic proteinuric nephropathy recruited in the COOPERATE study, combined RAS blockade by 3 mg trandolapril and 100 mg losartan daily was more effective than either drug prescribed alone at the same dose for reducing the incidence of the combined end-point of end-stage renal failure or doubling of serum creatinine concentration [29]. The situation is still more complex for congestive heart failure. On the one hand, it has been shown that, in contrast to the borderline benefit provided by high doses versus low doses of lisinopril in patients with congestive heart failure [30], the combination of an ACE inhibitor and an AT1R antagonist is more effective than the usual daily doses of either of these blockers given separately in patients with congestive heart failure (CHF) [31]. In the CHARM-added study, the addition of a mean daily dose of 24 mg candesartan to the usual doses of various ACE inhibitors, and to other recommended treatments for congestive heart failure such as beta-blockers and spironolactone, was more effective than a placebo in reducing cardiovascular deaths and the frequency of hospitalization for heart failure. On the contrary, in patients with myocardial infarction complicated by left ventricular systolic dysfunction, heart failure or both, and recruited in the VALIANT study, combination therapy with valsartan (80 mg b.i.d.) and captopril (50 mg t.i.d.) did not reduce mortality or the rates of key secondary outcomes, by comparison to captopril (50 mg t.i.d.) or valsartan (160 mg b.i.d.) [32]. A post-hoc analysis showed that combination therapy resulted in an apparent reduction in the cumulative rate of admission for recurrent myocardial infarction or heart failure. The VALIANT results confirm previous experimental data obtained in post-ischaemic heart failure in rats [33]. Thus, effective prevention of cardiovascular accidents, deaths and renal insufficiency by blockade of the RAS requires high doses of blockers rather than low doses, and the combination of two blockers increases blockade efficacy. Any binary classification into low and high doses is an oversimplification, arising from the dose-selection process based on decreases in blood pressure in hypertensive patients [34]. If an effect beyond that due to blood pressure reduction exists [35], then prescription of doses of ACE inhibitors or AT1R antagonists higher than those initially recommended to treat hypertension becomes a new objective. Because the combination of two blockers increases blockade efficacy more than a dose increase of each blocker, the role of combination therapy with AT1R receptor blocker and an ACE inhibitor is currently studied in a large trial involving patients with vascular disease [36]. The combination of two RAS blockers, acting at two different sites, should make it possible to minimize the daily doses required to obtain an ‘intense’ (in other words, complete and prolonged) RAS blockade. This approach is likely to decrease the incidence of some drug-related adverse effects, especially if the bradykinin-related adverse effects of ACE inhibition, cough and angioneurotic oedemia, are dose-dependent, which remains a matter of debate [37,38]. We already know that, in the VALIANT study a 160 mg b.i.d. dose of valsartan has been better tolerated than a 150 mg t.i.d. dose of captopril [32]. The prevention of the adverse effects directly related to the RAS blockade will require the same precautions, independently on the prescribed drug (AT1R antagonist or ACE inhibitor) or drug combination. Indeed, the term ‘intense’ RAS blockade provides no valuable information to help the physician to select the most appropriate daily dose for a particular drug for each patient. International normalized ratio (INR) and prothrombin time measurements are simple tests for the monitoring of anticoagulant treatment. No such simple means of practically quantifying RAS blockade is yet available to clinicians. Blood pressure is dependent on too many determinants and is difficult to measure. It is not feasible to measure decreases in vascular reactivity to exogenous angiotensin I and this factor seems to have no upper limit in response to increases in dose. In the long term, renin levels are adjusted by de novo renin synthesis [6]. N-Acetyl-seryl-arspartyl-lysyl-proline, a haematopoietic peptide, the metabolism of which is exclusively dependent on ACE activity and renal function, is easy to determine in urine and plasma, but the monitoring of this compound has not yet been tested on a large scale during long-term ACE inhibition; it is therefore difficult to determine whether this method could be used in practice to quantify inhibition and to monitor compliance [39,40]. In addition to a biomarker of ACE inhibition, we also urgently need a marker of the intensity of chronic RAS blockade, especially if renin inhibitors are to be introduced into human cardiovascular therapy [41]. Another issue concerns the risks of ‘reinforcing’ RAS blockade at the general population level. Experiments in rats and monkeys have shown that renin neutralization and the absence of sodium may be lethal [4,42,43]. In humans, we know that the same reinforced RAS blockade providing benefits in randomized controlled trials may be potentially dangerous if applied to the general population. In the CHARM-Added trial, 4.5% of the CHF patients receiving candesartan on top of an ACE inhibitor treatment experienced significant hypotension, necessitating treatment withdrawal, versus 3.1% of the patients in whom a placebo was added (P = 0.079) [31]. The incidence of a significant increase in creatinine concentration was almost twice as high in the candesartan-added group (7.8 versus 4.1% in the placebo-added group P = 0.0001), as was the incidence of hyperkalemia (3.4 versus 0.7%, respectively, P < 0.0001). In the VALIANT study [32], combination therapy had an additional blood pressure lowering effect but induced a clear increase in the rate of intolerance to treatment and permanent discontinuation of study treatment (9.0% versus 5.8% with valsartan and 7.7% with captopril). Adverse effects occasionally observed in controlled trials of reinforced RAS blockade, or of an anti-aldosterone therapy combined with ACE inhibition [44], may be serious and even lethal, in everyday practice. These adverse effects concern a minute fraction of the CHF patients treated, but this ‘minute fraction’ must be defined, as must ‘intense’ RAS blockade. Environmental aggression, such as acute dehydration, regional or general anaesthesia, onset or aggravation of CHF or dysrhythmia, may suddenly transform ‘healthy’ treated subjects into high-risk subjects, if RAS and aldosterone blockade are not rapidly ceased and/or if sodium repletion is not performed [45]. Combined blockade of the RAS implies the use of a wide range of doses and combinations to progressively initiate the treatment. It requires an educational message on the renin– angiotensin–aldosterone system, the sodium diet and the use of diuretics which is quite different from the standard advertising of ‘one-pill a day treatment to save lives’ [46]. After experimental and clinical investigation, pharmaco-epidemiological research will be required to assess, at the population level, the actual benefit/risk ratio of interrupting to various extents the vital physiological connection between the renin–angiotensin– aldosterone system, sodium and potassium balance, renal perfusion and blood pressure [47,48]. Acknowledgements We wish to thank the nurses of the HEGP Clinical Investigation Centre, Dr T. T. Guyene, Mrs M. F. Gonzalez and Mrs C. Dollin for their invaluable contribution to research in this area over the last 12 years. This research has been supported by grants from Institut National de la Santé et de la Recherche Médicale, Assistance Publique/Hôpitaux de Paris and Claude Bernard Association.