Individualised Selection of Antihypertensive Therapy
Post on 21-Mar-2017
Drugs 46 (Supp!. 2): 142-148, 1993 0012-6667/93/0200-0142/$3.50/0 Adis International Limited. All rights reserved.
Individualised Selection of Antihypertensive Therapy L.H. Opie Medical Research Council Ischaemic Heart Disease Research Unit, University of Cape Town Medical School, Cape Town, South Africa
Summary Theoretically, it should be possible to match the requirements of individual patients with the pharmacological and clinical properties of the large number of antihypertensive drugs now avail-able. The concept of automatic sequential stepped-care therapy is now largely outdated, but therapy of clinically important hypertension must be initiated with one agent. Diuretics remain a first-line option in the elderly and in Black patients, as do calcium antagonists. Outcome trials are available only for the elderly, and in these the benefits of initial diuretic therapy are well docu-mented. Nonetheless, diuretics may often need to be co-prescribed with a i3-blocker or an adren-ergic modifier such as methyldopa. i3-Blockers are preferred in patients with ischaemic heart disease or enhanced adrenergic drive, while a-blockers are preferred in patients with blood lipid abnormalities or prostatic problems. Calcium antagonists or angiotensin converting enzyme (ACE) inhibitors are being increasingly used as initial therapy when quality of life is important and metabolic neutrality is required.
Calcium antagonists are more likely to be effective first-line therapy than ACE inhibitors in patients with a high salt intake, in patients with Raynaud's disease, when angina pectoris is present, and in Black patients. ACE inhibitors are preferred for combination with diuretic agents, and in the presence of congestive heart failure or low salt intake. Experimentally, both calcium antagonists and ACE inhibitors can prevent ischaemic ventricular fibrillation and atheroma. Combination therapy between these 2 drug classes is gaining increasing acceptance because of these theoretical advantages. The clinical role of these agents, both singly and in combination, can only be fully evaluated by large scale outcome studies, which are now required.
Currently, the selection of antihypertensive therapy appears to be based on at least 5 factors. These include 1) experience (communal or indi-vidual), 2) comparative outcome studies, 3) sur-rogate end-points, 4) quality-of-life parameters and adverse effects of the agent used, and 5) indivi-dualised choice-care for the patient.
1. Clinical Experience
In the late 1970s, the Joint National American Committee on the Detection, Evaluation and Treatment of High Blood Pressure (1977) recom-
mended the use of thiazide diuretics as step 1 therapy for hypertension, with step 2 requiring the addition of an adrenergic modifier (propranolol, methyldopa, or reserpine). Step 3 required the ad-dition of the vasodilator hydralazine, and step 4 a ganglion-blocking agent such as guanethidine. None of these recommendations were based on long term comparative outcome studies, and therefore re-flected opinion and clinical practice rather than scientific evidence per se. The following year, the WHO Expert Committee on Arterial Hypertension (1978) suggested that either a ~-blocker or a di-uretic could be used as first-line therapy, with the
Individualised Selection of Antihypertensive Therapy
addition of a vasodilator to (j-blockade or addition of a (j-blocker or other adrenergic modifier to di-uretic therapy being recommended as the second-line approach. Again, these recommendations must have been based on clinical experience and enlight-ened guesswork. Hence previous clinical practice has been largely based on perceived wisdom rather than scientific data.
2. Comparative Outcome Studies
More recently, the publication of the first Medi-cal Research Council (MRC) trial (Medical Re-search Council Working Party 1985) made avail-able the results of the first large scale comparative study. In that study, both (j-blockers and diuretics were effective in the reduction of stroke, although diuretics appeared to be more effective. The cor-onary event rate was not reduced by diuretic therapy, but was reduced in nonsmokers receiving a (j-blocker. Thus, some glimmer of logic started to appear - diuretics, a simple treatment, reduced stroke very well, but did not alter coronary events. It should, however, be borne in mind that the di-uretic dose in that study was as high as was then the custom.
At the same time, Zanchetti (1985) proposed that diuretics, (j-blockers or angiotensin converting en-zyme (ACE) inhibitors be used as first-line therapy for hypertension. Two years later, Zanchetti (1987) had included calcium antagonists as alternative first-line agents. It was not until 1988 that the Joint National Committee (1988) followed the direction of Zanchetti and recommended a variety of agents as first-line treatments for hypertension. Even more recently, the Fifth Joint National Committee (1993) has bowed to logic and added a-blockers and a-(j-blockers as one of 6 possible alternative therapies.
Thus, what individual experience had estab-lished in the mid-1980s, i.e. that almost any anti-hypertensive drug could be used as fust-line treat-ment, has now been officially recommended. Nonethekss, it must be remembered that these rec-ommendations have not been based on large scale comparative outcome studies. The MRC trial did no more than compare 2 of the many possible first-
line therapies. Furthermore, almost all trials un-dertaken to date can be severely criticised. The overall impression given by the results from the 2 MRC trials, the first on middle-aged mildly hy-pertensive subjects (Medical Research Council 1985) and the second on elderly hypertensive sub-jects (Medical Research Council 1992), is that first-line treatment with diuretics produces a more fa-vourable outcome than first-line treatment with fJ-blockade.
In another comparative study, the Metoprolol Atherosclerosis Prevention in Hypertensives (MAPHY) Study, the (j-blocker metoprolol was compared with diuretic treatment over 4 years (Wikstrand et al. 1988). Metoprolol and a high dose of thiazide diuretic (hydrochlorothiazide 50mg) gave identical control of blood pressure, with lower total mortality in the group of patients receiving metoprolo1, because of fewer deaths from coronary heart disease and stroke. In contrast to the earlier MRC trial (Medical Research Council Working Party on Mild to Moderate Hypertension 1981), which used the (j-blocker propranolol, the MAPHY Study used a cardioselective (j-blocker and found the benefit of (j-blockade was largely found in smokers. The major problem with the MAPHY Study was the absence of a placebo group. There-fore, the apparent reduction in coronary mortality in patients receiving metopro101 could have been attributable to an excess mortality in the group not treated by metoprolol, possibly as a result of smok-ing and diuretic therapy.
3. Clinical Trials Evaluating Surrogate End-Points 3.1 Left Ventricular Hypertrophy
Now that left ventricular hypertrophy (L VH) has been identified as an independent risk factor for cardiova~ular mortality, interest in control of L VB has increased (Levy et al. 1990). Furthermore, the widespread availability of echocardiography with absolute quantification ofleft ventricular mass has also stirred interest in L VH as a surrogate end-point. In a ,meta-analysis of 109 studies, Dahlof et al. (1992a), demonstrated reversal of LVH with fJ-
blockers, calcium antagonists and ACE inhibitors, but found that the most effective agents were the ACE inhibitors. Diuretics also appeared to be ef-fective, but predominantly reduced ventricular di-ameter rather than ventricular wall thickness, thereby achieving only pseudo-regression of L VH.
In a subsequent prospective double-blind study, Dahlof et al. (1992b) examined the effects of 6 months' treatment with either enalapril or hydro-chlorothiazide. Patients who did not respond to monotherapy were excluded from the analysis. En-alapril reduced blood pressure chiefly through a re-duction in total peripheral resistance, while hydro-chlorothiazide reduced blood pressure primarily through a reduction in cardiac output. Left ven-tricular mass and wall thickness decreased more significantly in patients treated with enalapril than in those receiving hydrochlorothiazide. Fundal changes were significantly improved only in patients receiving enalapril. The results of this trial appear to suggest that reversal of cardiac hypertro-phy is related more closely to changes in the renin-angiotensin-aldosterone system than to the abso-lute blood pressure reduction.
3.2 Blood Lipid Patterns
There has been a single study comparing the effects of a diuretic, a l'1-blocker, an aI-blocker, a calcium antagonist, and an ACE inhibitor on blood lipid levels in patients with mild hypertension. In reality, patients included in this trial were virtually normotensive, with an average initial blood pres-sure of 140/91mm Hg. Many of the patients would not have fulfilled the primary criteria of many clin-icians for antihypertensive therapy. The decline in blood pressure in all treated groups was similar. Total cholesterol and high density lipoprotein (HDL) cholesterol tended to fall in all groups ex-cept in the group of patients receiving diuretic therapy. Compared with all the other groups, tne increase in total cholesterol with the diuretic chlor-thalidone was statistically significant (p < 0.01). This trial is of interest because the diuretic dose used was deliberately low (chlorthalidone 15 mg! day), confirming the clinical impression that di-
Drugs 46 (Suppl. 2) 1993
uretic therapy may adversely affect total blood cholesterol levels. Of the other drugs studied, the l'1-blocker acebutolol probably had the most bene-ficial effect on blood lipid levels. It is unclear whether acebutolol was beneficial because of its cardioselectivity or because of its intrinsic sym-pathomimetic activity, or both (Treatment of Mild Hypertension Research Group [TOMH] 1991).
3.3 Insulin Resistance and Carbohydrate Abnormalities
Although insulin resistance was first shown in a group of obese hypertensive patients who all had concomitant maturity-onset diabetes mellitus, it is now known that insulin resistance can also occur in nonobese, nondiabetic hypertensive subjects (Suzuki et al. 1992). Generally, it is thought that hyperinsulinaemia in essential hypertension may promote the severity of hypertension by increasing renal sodium reabsorption and/or by enhancing adrenergic drive. Some of the first studies in this area showed that diuretics reduced, whereas ACE inhibition improved, insulin sensitivity (Pollare et al. 1989). However, this benefit is not a specific property of ACE inhibition, and is also found with aI-blockade (Suzuki et al. 1992) and with calcium antagonism (Sheu et al. 1991). In ge~eral, l'1-block-ers appear to impair carbohydrate tolerance. For example, the effect of atenolol and nifedipine on glucose, insulin and lipid metabolism were com-pared in patients with hypertension (Sheu et al. 1991). In this study, plasma glucose levels did not change and plasma insulin levels rose in patients receiving atenolol, while in patients receiving treat-ment with nifedipine, plasma insulin levels were unchanged and plasma glucose and triglyceride levels fell.
Although these were only small changes, they support the general concept that ACE ~nhibitors, a-blockers, and calcium antagonists may be met-abolically neutral or even beneficial in patients with hypertension.
Individualised Selection of Antihypertensive Therapy
4. Quality-of-Life Parameters
Quality-of-life issues fIrst became important when the quality of life of patients treated with propranolol, methyldopa and captopril was com-pared (Croog et al. 1986). The quality of life with captopril was much better than was observed after treatment with the other 2 drugs, giving the impression that ACE inhibitors were particularly good at maintaining this parameter. Since then, however, several studies with various other anti-hypertensive agents, including calcium antagonists and (j-blockers other than propranolol, have shown that the quality of life was as good in drug-treated patients as it was in placebo-treated patients (re-viewed by Hjemdahl & Wiklund 1992). Diuretics also maintain patients' quality of life, except that a certain degree of sexual impairment appears to be a consistent fInding (Medical Research Council Working Party on Mild to Moderate Hypertension 1981; Treatment of Mild Hypertension Research Group, data on flle).
There have been several trials directly compar-ing calcium antagonists and ACE inhibitors (e.g. Os et al. 1991). Although the ACE inhibitor lisi-nopril caused fewer treatment withdrawals than ni-fedipine (presumably administered as a tablet, al-though this was not stated), the quality of life, as assessed by both patients and spouses, was very similar for both groups, except that a deterioration in quality of life was noted in patients treated with the highest dose of nifedipine (40mg twice daily).
In the TOMH study (Treatment of Mild Hyper-tension Research Group 1991), undertaken over a I-year period, doxazosin impaired quality of life slightly compared with a (j-blocker (acebutolol), a diuretic (chlorthalidone) or a calcium antagonist (amlodipine). a-Receptor blockade as initial therapy seems best tailored to the young, physically active patients in whom a low dose of a-blocker may be effective with minimal risk of adverse effects. In elderly men, the a-blocker prazosin may have the additional benefIcial effect of improving urinary flow in benign prostatic obstruction (Hedlun et al. 1983).
In summary, quality-of-life studies indicate that
propranolol specifIcally impairs quality of life and diuretics may predispose to sexual impotence, while most other antihypertensive agents leave the qual-ity of life unchanged (Hjemdahl & Wiklund 1992). As far as can be ascertained, the quality of life is similar whether patients receive ACE inhibitors or calcium antagonists.
5. Patient Selection Factors
The question of individualising antihypertens-ive therapy has recently been examined in detail (Opie 1992a). For each patient group, there are drug classes that should be most benefIcial as treatment of hypertension. For example, advantages of ACE
Table I. Assessment of ACE inhibitors compared with calcium
antagonists as treatment for hypertension (reproduced from Opie
Both drug classes are claimed to Maintain optimal phYSical, mental and sexual activity', with no
central adverse effects8
Maintain a lipid neutral profile
Leave insulin tolerance unchanged or possibly improved8
Reverse left ventricular hypertrophyb
Inhibit experimental atherosclerosisb
Improve arterial wall compliance8
Improve diabetic nephropathyC
Improve Raynaud's phenomenonb
ACE Inhibitors are preferred
In the presence of coexisting left ventricular systolic failure (calcium antagonists contraindicated)
In combination with diuretics When patients have a low salt intake
calcium antagonists are preferred
In combination with P-blockers (especially those calcium
,lntagonists that are dihydropyrldines)
In Black ethnic groups
In the presence of associated angina pectoris, especially
In severe urgent hypertension
In Raynaud's phenomenon
In pregnant patients
a Better documentation exists for ACE inhibitors.
b Better documentation exists for calcium antagonists.
c Better documentation exists for ACE inhibitors, although a recent comparative trial showed equal benefit for both groups
(Melbourne Diabetic Nephropathy Study Group 1991).
146 Drugs 46 (Suppl. 2) 1993
Table II. Experimental properties of calcium antagonists and ACE inhibitors as cardioprotective agents
Reference Drug group Property
Fleckenstein et al. (1989)
Opie (1990) Opie et al. (1988)
Muller et al' (1992)
Inhibition of ventricular fibrillation
inhibitors and calcium antagonists are shown in table I.
6. Comments on the Principles of Drug Selection
With so many principles governing drug selec-tion and so few outcome or surrogate end-point trials, it is not easy to provide definitive therapeu-tic recommendations for drug selection. It is clear that diuretics have been effective in studies un-dertaken in elderly patients and that they should be considered first-line treatment in these patients (Opie 1992b). The only comparative study in the elderly, which compared fj-blockers with diuretics, was marred by a high crossover rate, and 25% of the patients could not be followed up. Although it may be assumed that a major benefit in blood pres-sure treatment is through blood pressure reduction, it appears from the results of a large meta-analysis that other factors are also important (Collins et al. 1990). The most attractive hypothesis to explain the failure of the coronary event rate to fall with blood pressure is that the commonly used anti-hypertensive agents, namely diuretics and fj-block-ers, do in fact cause a series of metabolic impair-ments. These in tum might promote vascular disease. Thus, in patients other than the elderly, there is an increasing trend to use calcium antag-onists, ACE inhibitors or a-blockers as first-line therapy.
7. Selection of Combination Therapy
Almost all hypertensive patients have specific individual requirements that are best met by 'choice-care' .
It is disappointing that, as a general rule, any single agent will work in only about 50 to 60% of patients with mild to moderate hypertension when average doses are used. Even with higher doses the reported response rates are often not much more than 70%. As these higher doses lead to more fre-quent adverse effects, combinations of 2 or more agents, given in relatively low doses and acting by differen(mechanisms, are being used more often.
The following points were made by Opie (1992a). In general, ACE inhibitors are most effec-tive when combined with both a diuretic and a low salt diet (Singer et al. 1991). Furthermore, the com-bination of an ACE inhibitor with a diuretic is par-ticularly good therapy when a hypertensive patient also has congestive heart failure.
fj-Blockade combines well with calcium antag-onism because the former limits cardiovascular performance and decreases cardiac output, whereas the latter vasodilates and increases cardiac output. Nifedipine, in particular, combines well with fj-blockade because nifedipine has no effect on the sinus node or on cardiac conduction. An a-blocker may be combined with a fj-blocker or a diuretic when preservation of the normal blood lipid pro-file is important. When an a-blocker is combined with calcium antagonists, the peripheral vasodila-tion may be excessive, necessitating careful titra-tion of initial doses (J ee & Opie 1983).
Diuretics do not provide an added antihyper-tensive effect when combined with nifedipine-like calcium antagonists in mild to moderate hyperten-sion (Cappuccio et al. 1991; Salvetti et al. 1991). With verapamil, there is some evidence that the addition of diuretic therapy does improve anti-
Individualised Selection of Antihypertensive Therapy
hypertensive efficacy (Clement 1989; Holzgreve et al. 1989). This difference between the dihydropyr-idines and verapamil is in accordance with the much greater diuretic effect observed with nifedi-pine (10mg capsule) compared with that of vera-pamil (160mg tablet) [Leonetti et al. 1982].
The combination attracting increasing interest, and under preliminary investigation, is that of a calcium antagonist and an ACE inhibitor. The dif-ficulty of choosing between these 2 agents as initial therapy may be solved by using them in combin-ation. Both agents experimentally inhibit the de-velopment of ischaemic ventricular fibrillation and atheroma (table II). Hence, there are potential ar-guments for the prophylactic benefits associated with a combination exceeding the benefit of simple blood pressure reduction.
Capuccio FP, Markandu ND, Singer DRJ, et al. A double-blind crossover study of the effect of concomitant diuretic therapy in hypertensive patients treated with arnlodipine. American Journal of Hypertension 4: 297-302,1991
Clement DL. Calcium antagonists in the treatment of essential hypertension: review of international studies. JoumaJ of Car-diovascular Pharmacology 13 (Suppl. 4): 7-11, 1989
Collins R, Peto R, MacMahon S, et al. Blood pressure, stroke, and coronary heart disease. Part 2. Short-term reductions in blood pressure: overview of randomised drug trials in their epidemiological context. Lancet 335: 837-838, 1990
Croog SH, Levine S, Testa MA, et al. The effect of antihyperten-sive therapy on the quality of life. New England Journal of Medicine 314: 1657-1664, 1986
Dahlof B, Pennert K, Hansson L. Reversal ofleft ventricular hy-pertrophy in hypertensive patients. A meta-analysis of 109 treatment studies. American JoumaJ of Hypertension 5: 95-110, 1992a
Dahlof B, Herlitz H, Aurell M, Hansson L. Reversal of cardio-vascular structural changes when treating essential hyperten-sion. The importance of the renin-angiotensin-aldosterone sys-tem. American JoumaJ of Hypertension 5: 900-911, 1992b
Fleckenstein A, Fleckenstein-Griin G, Frey M, Zorn J. Calcium antagonism and ACE inhibition. Two outstandingly effective means of interference with cardiovascular calcium overload, high blood pressure, and arteriosclerosis in spontaneously hy-pertensive rats. American JoumaJ of Hypertension 2: 194-204, 1989
Hed1un H, Andersson KE, Ek A. Effects of prazosin in patients with benign prostatic obstruction. JoumaJ of Urology 130: 275-278, 1983
Hjemdahl P, Wiklund IK. Quality of life on antihypertensive drug therapy: scientific end-point or marketing exercise? JoumaJ of Hypertension 10: 1437-1446, 1992
Holzgreve H, Distler A, Michaelis J, Philipp T, Wellek S, on be-half of the Verapamil versus Diuretic (VERDI) Trial Research
Group. Verapamil versus hydrochlorothiazide in the treatment of hypertension: results onong-term double-blind comparative trial. British Medical JoumaJ 299: 881-886, 1989
Jee LD, Opie LH. Acute hypotensive response to nifedipine added to prazosin in treatment of hypertension. British Medical Jour-nal 287: 1514-1516, 1983
Joint National American Committee on the Detection, Evalua-tion and Treatment of High Blood Pressure. National Institute of Health, Bethesda, 1977
Joint National Committee. The 1988 report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure. Archives ofIntemaJ Medicine 148: 1023-1038, 1988
Joint National Committee on Detection, Evaluation, and Treat-ment of High Blood Pressure. journal/publisher tic, in press, 1993
Leonetti G, Cuspidi C, Sarnpieri 1, et al. Comparison of cardio-vascular, renal, and humoral effects of acute administration of two calcium channel blockers in normotensive and hyperten-sive subjects. JoumaJ of Cardiovascular Pharmacology 4: S319-S324, 1982
Levy D, Garrison RJ, Savage DD, et al. Prognostic implications of echocardiographicaJly determined left ventricular mass in the Framingham Heart Study. New England Journal of Med-icine 322: 1561-1566, 1990
Medical Research Council Working Party on Mild to Moderate Hypertension. Adverse reactions to bendrofluazide and pro-pranolol for the treatment of mild hypertension. Lancet 2: 539-543, 1981
Medical Research Council Working Party. MRC trial of treat-ment of mild hypertension: principal results. British Medical JoumaJ 291: 97-104, 1985
Medical Research Council triaJ of treatment of hypertension in older adults: principal results. British Medical JoumaJ 304: 405-412, 1992
Melbourne Diabetic Nephropathy Study Group. Comparison be-tween perindopril and nifedipine in hypertensive and normo-tensive diabetic patients with microalbuminuria. British Medi-cal JoumaJ 302: 210-216, 1991
Muller CA, Opie LH, Peisach M, Pineda CA. Antiarrhythmic ef-fects of the angiotensin converting enzyme inhibitor perin-doprilat in a pig model of acute regional myocardial ischaemia. JoumaJ of Cardiovascular Pharmacology 19: 748-754, 1992
Opie LH. Calcium channel antagonists. Part VIII. Current de-velopments: calcium antagonists in primary and secondary prevention of coronary heart disease. In Clinical use of cal-cium channel antagonist drugs, 2nd ed., pp. 286-307, Kluwer Academic Publishers, Boston, 1990
Opie LH. Choosing the cOrrect drug for the individual hyperten-sive patient. Drugs 44 (Suppl. 1): 147-155, 1992a
Opie LH. Treatment of hypertension in the elderly - the end of the story? Editorial. Cardiovascular Drugs and Therapy 6: 559-562, 1992b
Opie LH, Coetzee W A, Dennis SC, Thandroyen Fr. A potential role of calcium ions in early ischemic and reperfusion arrhyth-mias. Annals of the New York Academy of Sciences 522: 464-477, 1988
Os I, Brat1and B, Dahlof B, et al. Lisinopril or nifedipine in es-sential hypertension? A Norwegian multicenter study on effi-cacy, tolerability and quality of life in 828 patients. JoumaJ of Hypertension 9: 1097-1104, 1991
Pollare T, Lithel H, Berne C. A comparison of the effects of hydrochlorothiazide and captopril on glucose and lipid me-tabolism in patients with hypertension. New England Journal of Medicine 321: 868-873, 1989
Salvetti A, Magagna A, Innocenti P, et al. The combination of chlorthaJidone with nifedipine does not exert an additive anti-hypertensive effect in essential hypertensives: a crossover mul-
ticenter study. Journal of Cardiovascular Pharmacology 17: 332-335, 1991
Sheu WHH, Swislocki ALM, Hoffman B, et al. Comparison of the effects of atenolol and nifedipine on glucose, insulin, and lipid metabolism in patients with hypertension. American Journal of Hypertension 4: 199-205, 1991
Singer DRJ, Markandu ND, Sugden AL, et al. Sodium restriction in hypertensive patients treated with a converting enzyme in-hibitor and a thiazide. Hypertension 17: 798-803, 1991
Suzuki M, Hirose J, Asakura Y, et al. Insulin insensitivity in non-obese, non-diabetic essential hypertension and its improve-ment by an alpha I-blocker (bunazosin). American Journal of Hypertension 5: 869-874, 1992
Treatment of Mild Hypertension Research Group (TOMH). The treatment of mild hypertension study. A randomized, placebo-controlled trial of a nutritional-hygienic regimen along with various drug monotherapies. Archives ofInternal Medicine 151: 1413-1423, 1991
WHO Expert Committee. Arterial hypertension. Technical Re-port Series 628, World Health Organization" Geneva, 1978
Wikstrand J, Warnold I, Olsson G, et al. Primary prevention with metoprolol in patients with hypertension. Morlality results from the MAPHY study. Journal of the American Medical Asso-ciation 259: 1976-1982, 1988
Zanchetti A. A re-examination of stepped-care: a retrospective and a prospective. Journal of Cardiovascular Pharmacology 7: SI26-S131, 1985
Zanchetti A. A role of calcium antagonists in systemic hyperten-sion. American Journal of Cardiology 59: 130B-136B, 1987
Correspondence and reprints: Prof. L.H. Opie, Medical Research Council Ischaemic Heart Disease Research Unit, University of Cape Town Medical School, Cape Town, South Africa.
Discussion Prof Z. Rumboldt (Croatia): Professor Opie, you
discussed everything in your presentation except the price of therapy. Diuretic therapy is generally between 1 and 2 orders of magnitude cheaper than other antihypertensive therapy.
Prof L. Opie: I believe that the arguments for low dose diuretic therapy are strong. However, di-
Drugs 46 (Suppl. 2) 1993
uretics would not be the most appropriate therapy for many patients (e.g. those with diabetes or with lipid abnormalities).
Prof A. Zanchetti: Could I comment on the use of calcium antagonists in congestive heart failure? You suggested that the more vascular-selective agents may be more useful clinically because they are potentially less negatively inotropic and cause less tachycardia. I would like to suggest a different explanation. Of the drugs you discussed, amlodi-pine is poorly selective; in fact, it is less selective than nifedipine. It may be that fast-acting dihy-dropyridines cause excessive sympathetic activa-tion in conditions such as congestive heart failure. Therefore, these agents may be less useful clinically not because they are negatively inotropic but be-cause some of them stimulate the sympathetic sys-tem excessively.
Prof Opie: Thank you for your comment Pro-fessor Zanchetti. Perhaps I did emphasise that you cannot look at the list drugs and say that the dif-ferences are explained by vascular selectivity; ni-soldipine is actually 1000 times more vascular se-lective than the others and, as you say, amlodipine is similar to nifedipine. Another important factor is the duration of activity; amlodipine is very long acting and in those studies nisoldipine was used in a very short-acting preparation. It appears that in-termittent activation of the renin-angiotensin ad-renergic system causes unfavourable effects that are best avoided by using a long-acting compound. Finally, in the amlodipine study ACE inhibitors were generally coadministered.